|
My master thesis, entitled "Happiness and Economics:
Enriching economic theory with empirical psychology", was submitted to
Maastricht University on Sep. 27, 2001.
About my thesis “Happiness
and Economics”
“Happiness and Economics” investigates the relationship between choice,
material welfare, and happiness. Empirical evidence on subjective
well-being is analyzed and set against economic theory. It appears that
strong a priori assumptions limit the economic science in its ability to
provide substantive explanations of human behavior. As a consequence,
economics has lost sight of its original concern for genuine well-being.
Instead, economists take delight in the internal consistency and
analytical versatility of their theoretical models.
Themes discussed in “Happiness and Economics” include
- rational behavior: freedom of will vs. determinism (ch. 1, part 1)
- the moral dimension to Pareto efficiency (ibid.)
- subjectivity vs. objectivity in scientific research (ch. 1, part 2)
- validity of happiness surveys (ibid.)
- cardinality and interpersonal comparability of utility (ch. 1)
- the nature of happiness (ch. 2, part 2)
- economics and (lack of) empiricism (ch. 1 and 3)
- the potential harms and benefits of economic growth (ch. 3)
- the relation between money and happiness (ch. 2 and 3)
- implications for economic theory and policy (ch. 3)
- cultural relativism with respect to happiness (ch. 2 and 3).
Download
You can download the full text in Adobe Reader (*.pdf) format:
» for screen display (1.2 MB, includes hyperlinks)
» for printing on A4 paper (1.6 MB, for double-sided printing)
» for printing on US letter paper (1.6 MB, for double-sided printing)
Excerpts
select a topic from this list
introduction
the economic conception of behavior
subjectivity and scientific rigor
happiness and life circumstances
income and happiness
happiness as a judgment
individual rationality can be socially wasteful
happiness research thwarts the rationale for
social discounting
the scope of happiness research
the procedure can matter more than the outcome
conclusion
[page numbers refer to the print version]
introduction
“Despite decades of economic growth and unparalleled technological
progress, it seems that people in the wealthy countries have done badly
in transforming material affluence into freedom from economic cares
and into agreeable lives. While there can be little doubt that in many
respects life in the industrialized countries is better today than in
1928, it appears surprising that the economic problem is far from being
solved. Material affluence—American per capita income rose by 277%,
i.e., almost four-fold, between 1913 and 1989 (Maddison 1991)—it seems,
has done little to bring real improvements in well-being. More than
half of Americans (since 1973, when the series starts) agree with the
statement that “the lot of the average man is getting worse”, and the
share of people agreeing has been rising after 1981 (until 1994 where
the series stops) to reach almost two thirds (Lane 2000b: 27-8). Indeed,
as will be shown later, self-reported life satisfaction in the U.S.
has, if anything, shown a downward trend after World War II while it
did not show any clear trend for the industrialized world as a whole.”
(Introduction, p. 5)
back to list
the economic conception of behavior
“The behavioral theory that has been developed within economics has
not been inspired by psychological research or any other sort of empirically
qualified concepts. Rather, it relies primarily on a priori axioms that
derive their appeal from their theoretical properties. In particular,
economics has adopted a conception of behavior that is largely identical
with the Hobbesian doctrine of psychological hedonism that states that
nature endowed the human mind with but a single motivation which is
the attainment of pleasure. Even ostensibly altruistic acts are deemed
to be motivated by the ensuing pleasure one experiences. Thus Thomas
Hobbes (1588-1679) is said to have explained, after giving alms to a
beggar outside St. Paul’s Cathedral, that he gave the money because
it pleased him to see the poor man pleased.
There has certainly not been a lack of criticism of the Hobbesian view,
but as long as the scientific community does not agree on ways to observe
human motivation (and even introspection seems to be a questionable
standard), every complete account of human motivation must remain beyond
falsification. Consequently, that also means that psychological hedonism
cannot be a scientific theory by the well-known criteria established
by Karl Popper (1959/1934).” (chapter 1, p. 7)
back to list
“When now the rational behavior hypothesis is merged with the utility
maximization axiom, the economic conception of human behavior is complete.
Consistent with the terminology used thus far, this conception of behavior
will be called the rational utility maximization hypothesis (RUMH
for brevity; panel V in fig. 2). To summarize briefly, the RUMH contains
the axioms and hypotheses that
- Ultimately, an individual’s values boil down to one summum bonum
which is called, by convention, utility.
- Motives completely determine behavior.
- Utility maximization is the only motive.
- People make reasonably precise predictions of the consequences of
their choices.
As a consequence, and this is the most important result of this analysis,
the RUMH states that people reasonably well maximize their experienced
utility. [...]
The exclusion of a free will therefore leads to a loss of distinction
between values and motives and necessarily implies hypothesis 3. This
is a striking result worth to appreciate for a moment. As soon as behavior
is assumed to be determined, the utility maximization axiom follows
necessarily, unless gross inconsistencies are committed. The single
assumption of given values, or, in economic terminology, of a connected
preference map leads to the result that all behavior can be interpreted
as motivated by utility maximization. Put differently, the entire idea
of utility maximization follows rather automatically once a free will
is excluded.” (chapter 1, pp. 12)
back to list
subjectivity and scientific rigor
“Thus, subjectivity is not a weakness per se. On the contrary, when
we are concerned with theories concerning the human mind, we cannot
do without subjectivity. And when economics wants to give substantive
explanations of behavior, it must be concerned with the human mind,
not only in normative economics but also in its positive branch as far
as it is concerned with choice. Objectivity might often be of instrumental
value in order to elucidate partial relationships and mechanisms, but
it is no compelling criterion. Rather, the ultimate criterion has to
be the relevance to the problem at hand and the scientific rigor. Scientific
rigor, in turn, is not dependent on objectivity and precision, but primarily
requires concreteness to warrant falsifiability.” (chapter 1,
p. 22)
back to list
happiness and life circumstances
“If we step back from single correlates, it appears that “hard” (i.e.,
readily measurable) objective life circumstances as a whole account
only for a small proportion of the observed variance in SWB, and that
the large untraceable proportion of variance among individuals must
be attributed to other than these factors. Taking all their demographic
factors together, Campbell, Converse, and Rodgers (1976) could explain
not even 20% of the variance in SWB (reported by Ed Diener and Lucas
1999). Andrews and Withey (1976: 141) could not explain more than 11%
in any of their data sets when they included six demographic variables
simultaneously (family life cycle stage, family income, age, education,
race, sex), and none of these predictors alone could explain more than
6% of the variance. These low values suggest that “there is no reason
to expect strong relationships between the objective conditions of life
and subjective assessments of well-being under most circumstances” (Schwarz
and Strack 1999: 79).” (chapter 2, p. 33)
back to list
income and happiness
“The above presentation of income correlates with SWB can be summarized
by three stylized facts.
-
At one point in time, within a given country, the poor report lower
SWB than the non-poor.
-
Over time, within a given country whose per capita income exceeds
some threshold income, SWB does not rise as per capita income rises.
-
Across countries, happiness correlates positively with per capita
income, but the relationship is weak to nonexistent among the richer
countries. [...]
The first and second stylized facts suggest that there is a considerable
relative aspect to the SWB-income relationship: people derive satisfaction
of earning an income that is high with respect to the country’s current
average income, but apart from this relative effect the level of income
has no bearing on happiness as long as it exceeds some threshold income.
This observation has become generalized as the relative income hypothesis.”
(chapter 2, p. 36)
back to list
“In even the poorest countries the consumption patterns of the world’s
wealthiest societies are represented, in particular in the cities, via
television, tourism, advertising, and through the emulation of Western
life styles by the local upper classes. Lauterbach (1972), for example,
observed that in Latin America the exposure to the consumption patterns
of the affluent countries gives rise to “‘premature’ consumer aspirations”
and to “an ever-growing pressure on the Latin American consumer, even
the one with little discretionary purchasing power, in the direction
of emulating the consumption preferences and habits of consumers in
industrialized countries” (ibid.: 276).” (chapter 2, p. 42)
back to list
“However, taking into account all the evidence considered so far, a
preliminary conclusion on the relationship between income and SWB shall
be drawn.
A priori, there can be no doubt that for very low levels of
income there will be a positive and causal influence of income on SWB
that does not depend on the income of any other individual. Once a certain
threshold income is reached, however, a nation’s average SWB does not
respond to increases in income while the SWB of any individual tends
to increase with relative income. Despite the positive cross-national
correlation between income and SWB, there seems to be no substantial
causal relationship among these two variables. Rather, the correlation
seems to be primarily, if not exclusively, due to a third influence,
that of individualism (and perhaps additional cultural attributes).
One effect of individualism is to fuel competition which results in
economic efficiency and, therefore, in material prosperity. A second
effect is that individualism places priority on intrinsic desires which
are characterized by a high “happiness payoff”. At the same time, it
assigns a more important role to happiness as a constituent of the good
life.” (chapter 2, p. 44)
back to list
happiness as a judgment
“Very generally, a judgment can be understood as an evaluation of a
state against a counterfactual standard of what ought to be. This standard,
which I will call the counterfactual frame of reference, will
be influenced basically by two hypothetical considerations: by the perceived
likelihood of alternative states (likelihood counterfactuals)
and by normative considerations of what ought to be the case (normative
counterfactuals; Kahneman and Miller’s [1986] norm theory
takes the same perspective). [...]
The entire question of SWB, and consequently the very idea of the good
society, appears in a paradigmatically different light if it is not
understood as maximizing the availability of resources that meet given
needs, but as the satisfaction of legitimate expectations in which the
active and sensible examination of the expectations themselves has at
least as large a role to play as the means to satisfy them. Csikszentmihalyi,
renowned for his seminal research on truly gratifying experiences he
called “flow”, expressed in one sentence all the difference such a perspective
makes: “Happiness is not something that happens to people but something
that they make happen” (Csikszentmihalyi 1999).” (chapter 2, pp. 46)
back to list
individual rationality can be socially wasteful
“To the individual, the desire for positional goods, or for status
itself, is an aspiration not categorically different from the desire
for a nice house or good health. From the viewpoint of society, however,
the satisfaction of positional aspirations is a zero-sum game without
a net benefit: the gain of one is, by definition, the loss of another.
Only to the degree that those with above-average aspirations “purchase”
high positions from those with below-average aspirations (e.g., as in
the case of within-firm status-salary bargains; see chapter two, p.
39) a net social gain is possible. Once the distribution of positions
has exhausted such potential gains, however, and in the majority of
settings where prohibitive transaction costs prevent such gains altogether,
the individually rational pursuit of status will lead to a waste of
resources from a social perspective.” (chapter 3, p. 52)
back to list
happiness research thwarts the rationale
for social discounting
“In this context it should be noted that a major argument for the economic
practice of discounting future outcomes would not be valid in a SWB
perspective. If wealth is understood in terms of goods, discounting
over generations can be justified on the grounds that the discount rate
reflects the potential increase in real wealth. For example, at a social
discount rate of three percent and assuming no inflation, environmental
damage one hundred years hence of $1 million will have a net present
value of $52,033. If avoiding the damage requires an investment today
of more than this amount, such an investment will be inefficient because
investing it elsewhere will lead to a benefit that will in one hundred
years exceed the cost of the environmental degradation, leaving a net
benefit to society. However, the analysis will be invalid in terms of
SWB. The failure of SWB, in the wealthy countries, to rise over time
despite economic growth undermines the very rationale for intergenerational
discounting. No purely quantitative method will provide an appropriate
criterion for long-term tradeoffs between different forms of investment.”
(chapter 3, p. 62)
back to list
the scope of happiness research
“It should be clear therefore that the goal of happiness research cannot
be to provide the ultimate criterion for the good life and the good
society. Rather, its goal has to be understood as contributing to a
better understanding of one, perhaps the most important, aspect of individual
and social welfare. Given that happiness is one among several values
people hold, measuring SWB can help identify those factors that add
to happiness—and those that do not. Perhaps its greatest value lies
in testing and correcting intuitive theories of what affects SWB.“ (chapter
3, p. 62)
back to list
the procedure can matter more than the outcome
“The way of thinking about economic policy should receive a different
emphasis, away from the traditionally dominant concern for outcomes
towards a concern for procedures. Solid evidence from laboratory and
field experiments shows that procedural utility, i.e., considerations
of procedural characteristics independent of the actual outcome, has
a considerable and independent influence on individuals’ satisfaction
(Folger 1986; Frey and Stutzer 1999b, 2001a; Tyler 1990). Just as Richard
Layard is more annoyed by an identical noise at night if it is due to
his neighbors than to the wind (see p. 47), people in general have been
found to evaluate identical outcomes very differently on the basis of
the circumstances and procedures that brought them about. For economic
policy this means that the SWB of consumers, who are always also voters,
workers, tax payers, and citizens, may depend more on the design of
institutions and possibilities of participation than on consumption.
These themes are rightly emphasized by Constitutional Political Economy.
Swiss authors in particular—before the background of a long tradition
of federal, participatory democracy, but also of Europe’s highest incomes—point
to the benefits of federalism and direct democracy and provide econometric
evidence that these are related to sound fiscal policies and, indeed,
happiness (Frey and Stutzer 1999a, 1999b, 2001b). In turn, greater satisfaction
with political institutions, and with life in general, may feed back
to economic prosperity. A citizen who feels that she co-determines how
much taxes she pays may be less likely to feel discouraged by a given
marginal tax rate than one who feels taxes are imposed by an independent,
anonymous authority.” (chapter 3, pp. 62)
back to list
conclusion
“Economics should take advantage of more than
forty years of empirical SWB research. The recognition that self-report
measures are valid and reliable and that they correspond precisely to
the subjective spirit of the psychological premises of economics would
enrich the economic science in two ways: by substantiating the economic
concept of utility and clarifying the limits of economics.
A substantiation of the economic
utility concept would be a great progress for the economic science at
large. It would render economic theory falsifiable and thereby add value
to economics analyses because only falsifiability ultimately warrants
a growth of knowledge (Popper 1959/1934). Theories based on unfalsifiable
premises of what motivations underlie human behavior belong, in the
sense of Popper, to the realm of metaphysics. Since their propositions
cannot be refuted, they can also not be corroborated.
Making use of empirical methods
to measure utility would enable economics to develop more substantive
theories of how the behavior of individuals adds up to aggregate outcomes.
The objection that the substantiation of the utility concept would overcharge
economics with redundant details will be invalid. On the one hand, as
has been indicated, taking SWB seriously will in many contexts lead
to a substantial improvement of the explanatory potential of economics.
It will therefore not be redundant but, on the contrary, indispensable.
On the other hand, the current thin concept of utility will continue
to be a powerful heuristic device that can inspire economists in their
endeavor to explain economic phenomena. It will also be a useful empirical
approximation of behavior in many more narrowly circumscribed contexts.
[...]
The single most significant lesson
of SWB research for economics is perhaps the recognition that negative
externalities of individual choices are not the exception but the rule.
As a consequence, the almost unconditional trust of many economists
in the invisible hand is no longer justified, even though the concept
itself will not become less valid. Putting the invisible hand in due
perspective will reveal that there is a long way to go from individual
choice to social welfare. Furthermore, it will necessitate a shift of
emphasis from quantitative to qualitative wealth. Without institutional
arrangements that restrict positional competition, a society is likely
to spend too many resources on jewelry and advertising, to work too
many hours, and to teach their children to take positional competition
too seriously. An economy may produce quantitatively much in terms of
goods, but in how far these goods translate into high well-being depends
on how wisely a society sets the rules of the game and how it educates
its children.
The pervasiveness of externalities
also qualifies the Pareto criterion. Above all, it becomes apparent
that the Pareto criterion derives its intuitive appeal from its supposed
legitimacy. Its wide acceptance rests on the conviction that it tolerates
the tolerable, namely personal tastes that do not unduly interfere with
the liberty of others, and does not tolerate the intolerable, namely
envy and other tastes that do interfere unduly with the liberty of others.
Once it is recognized, however, that externalities are pervasive; that
these externalities persist even in the absence of transaction costs;
and that externalities from positional competition are not as easily
labeled ‘legitimate’ or ‘illegitimate’ as conventional externalities
(like industrial pollution), it becomes clear that the criterion of
efficiency depends ultimately on which preferences and which degree
of interference with the liberty of others are considered legitimate.
In other words, it is ultimately absolute values, and not relative preferences,
that define what is efficiency, individual well-being, and social welfare.”
(conclusion, p. 66)
back
to list
|
