[Note for bibliographic reference: Melberg, Hans O. (1997), What is economics? A
method or a topic?, www.oocities.org/hmelberg/papers/971018.htm]
What is economics?
A method or a topic?
by Hans O. Melberg
Is it proper to publish a paper on suicide in an economic journal? Should economists avoid
explanations (e.g. of unemployment and growth) based on norms and culture? Should
economists do empirical work on the mechanisms affecting the formation of beliefs and
preferences?
The answer to these questions depend on our conception of economics. If the essence of
economics is the analysis of behaviour on the assumption that it is inspired by rational
cost/benefit considerations, then even suicide might be within the realm of economics
(suicide occurs when the perceived cost of living is lower than the perceived benefits).
However, on this view of economics, it would be wrong to bring in norms when explaining
economic phenomena. All economic explanations have to be based on the assumption that
behaviour is rational. On the other hand, if economics is considered to be the analysis of
a set of issues - unemployment, economic growth, inflation, consumption, investment and so
on - then norms could well turn out to be an important explanatory variable.
So, we have at least two rival concepts of economics. The first is economics as an
approach - a methodology. An current example of an economist in this tradition, is Gary
Becker. Building on "the economic approach to human behaviour" he has used
cost/benefit analysis to explain crime, the decision to have children, the choice of
marriage partner, divorce, and many other non-standard topics for mainstream economists.
The second approach, economics as a set of topics - is implicit in R. Solow's 1980
Presidential Address about unemployment, in which norms about fairness was emphasized as a
major cause of wage rigidities and hence unemployment. An explicit proponent of the same
approach, is R. Coase who writes about "my belief that in the long run it is the
subject matter, the kind of question which the practitioners are trying to answer, which
tends to be the dominant factor producing the cohesive force that makes a group of
scholars a recognisable profession with its own university departments, journals and
libraries" (Coase, 1994, p. 38).
I could go on giving examples of economists who belong to one of the two traditions.
Steven E. Landsburg, for instance, belongs very strongly to the
"economics-as-an-approach-camp." When faced with apparently irrational phenomena
like $2.99 pricing (why not $3?), long queues to get tickets to a concert (why not rise
the price?), and voting (seldom worth the trouble on a pure cost/benefit consideration) he
"insists on finding rational explanation, no matter how outlandish, for all of this
apparently irrational behaviour" (1993, p. 11-12). Douglas C. North, on the other
hand, belongs to the camp who wants to incorporate norms, culture and institutions into
economics, as is evident from his writings about the causes of economic growth.
So, who is right? To answer this, I shall examine two reasons for the somewhat dogmatic
adherence to the assumption of rationality.
First, some might claim that irrational and non-rational behaviour (behaviour guided by
norms and passions) are not very important. Hence, there is little point in spending a lot
of time on a topic which is not likely to explain many phenomena. And, even if the
explanatory capacity of our theories could be improved by incorporating non-rational
behaviour, we might argue that the gains are illusory. Bringing in new variables always
increase our explanatory capacity, as we know from statistical theory (R2 always goes down
when we introduce more variables). What we want is parsimonious explanations: to explain
as much as possible using as little as possible (Occam's razor). The only way to convince
these economists, would be to give concrete empirical evidence demonstrating that norms
and passions are important causes of behaviour.
Second, even if one admits that irrationality is not of marginal importance, one could
argue that the marginal payoff from spending time on the topic is low. The payoff has two
dimensions. First, the increase in the amount of phenomena we can claim to explain.
Second, the change in the reliability of our explanations. Now, the traditional economist
would argue that bringing in norms might increase the scope of economics, but it might
also reduce the reliability of our theories. It reduces the reliability since the causes
of irrational phenomena are difficult to quantify. This makes it hard to test the truth of
explanations based on norms, and thus it reduces the reliability of our theories. Hence,
to show that norms and passions have a place in economics, we must demonstrate that it is
possible to quantify and test these theories.
These are valid and good reasons to insist on the assumption of rationality, and to
resist the sociologization of economics. It is, however, also falsifiable since its force
will be reduced if people can come up with concrete empirical studies demonstrating the
importance and reliability of explanations based on norms and passions.
There is, however, also a third option: To retain the supremacy of the assumption of
rationality (and the variables of beliefs and preferences), but to extend it to the
formation of belief and preferences. For instance, the assumption of rationality should
not only apply to the choice of means to achieve a given end (Elster's "thin"
concept of rationality). We might also examine whether the beliefs are formed rationally,
and - more controversially - whether preferences are formed rationally (Elster's
"broad" concept of rationality). Rational beliefs require optimal collection of
information and optimal processing of the information (no bias resulting from hot or cold
mechanisms such as wishful thinking and cognitive illusions). Rational preferences can be
defined as preferences that are not subverted by mechanisms which we, if we were aware of
the mechanisms, would not accept. For instance, if I do not want something simply because
I cannot get it (the mechanism of sour grapes), then one might question whether my
preference has been formed rationally. Using the broad theory of rationality, we end up
with what might be called "psychological economics"; a close focus on the
psychological mechanisms that actually shape our beliefs and preferences.
Of course, the traditional economists can object to this psychologization of economics.
For instance, he might claim that it is perfectly acceptable to take preferences and (less
likely) beliefs as given. In short, scarcity of time and space prevents economists from
discussing the causes of the variables they are using to explain. There is always a deeper
cause of the cause; behaviour can be explained by preferences and beliefs, preferences and
beliefs can be explained by psychology, psychology might be reduced to biology (genetics),
biology can be reduced to chemistry and so on [The regression might stop here (with
chemistry) - unless we include God. As the well known chemist P. Atkins writes:
"Because our brains are made up of elements, even our opinions are, in a sense,
properties of the elements and inhabitant of the [periodic] kingdom" (Atkins (1995),
p. 6)). Hence, to complain that economists do not explain preferences, is not a critique
with great force: it is perfectly valid to focus on one step in the causal chain; nobody
can discuss everything all the time!
Once again, the argument has some force. At least it presents a division of labour:
Economists deal with the consequences of choices given preferences and beliefs,
psychologists focus on the formation of preferences and beliefs etc.,. However, there are
also good arguments for at least going one step back in the causal chain. As Houthakker
writes: "There is much to be said for regarding the explanation of the content of
preferences ... as outside the economist's competence. Nevertheless, there is also a
danger in such delimitations of responsibility, namely that the problems thus excluded may
not be studied at all. It is easy to say they belong to psychology but this does not mean
that psychology will find them sufficiently interesting to look into them." (p. 156,
in Mason (1988)).
The danger of beliefs and preferences being left unexplained might be exaggerated, but
there are some beliefs and preferences that psychologist seldom focus on. Take, for
instance, the case of why people believe what they do about free trade. Economists are
probably more likely to wonder about this question since they hold opinions that differ
greatly from the general public.
As a side-note, one might note that "imperial economists" do not restrict
their explanatory claims to behaviour. Also beliefs and preferences can be explained by
cost/benefit calculations they claim. As for beliefs, this is a well established branch
(rational expectations), but preferences have so far been kept outside (De gustibus non
est disputandem). However, lately the economist Gary S. Becker has also tried to explain
preferences as the outcome of rational choice (see Accounting for Tastes). [Note:
It is perfectly possible to explain preferences rationally, at the same time that you do
not make a claim on the rationality of the preference. Whether the cause of a preference
is a rational choice is conceptually distinct (but related?) from whether the preferences
actually is rational (if such a question can be asked]. So, instead of talking about the
psychologization of economics, one might choose a fourth approach: the economization of
psychology!
A frame
My discussion so far can be summed up in a diagram:
What is economics?
Topic
Economic Behaviour Beliefs & Preferences Behaviour in general
..Rationality Traditional economics Economic psychology Imperial economics
Method
..Norms Sociological economics Psychological economics Traditional sociology
In short, traditional economics tries to answer what is considered to be standard
economic questions using the assumption of rationality. The assumption of rationality has
also been used to explain beliefs, and lately some economists have taken an even more
controversial step and used rational choice theory to explain preferences. I have labelled
this approach "Economic psychology". Another development has been the invasion
of economists into questions previously though to be outside the realm of economics
(interaction in the family, suicide, crime) to produce what is sometimes called
"imperial economics", or less pejorative, economic sociology (also: economic
political science, such as Public Choice). Some economists, however, complain that the
assumption of rationality is not an adequate description of behaviour even on topics
considered to be traditional economics (growth, unemployment). To give good answer to
economic questions, they argue, we have to bring in norms, culture, passions and
institutions. In short, we need more "sociological economics." A related
complaint, is that economic phenomena cannot be adequately explained as long as we do not
study more closely exactly how people form their beliefs and preferences (not just assume
it is done rationally). The study of how non-rational beliefs and preferences are formed
and affect the answer to economic question, are labelled "Psychological
economics." Finally, there is traditional sociology which tries to explain
non-economic topics using norms and other non-rational motivational factors as their main
explanatory device.
Now, to ask who is right and where economists should be in my grid, only gives meaning
in those cases in which there is a conflict and we have a neutral criteria. There is
sometimes a conflict between rational and non-rational explanations of behaviour; but it
is not mutually exclusive in the sense that all behaviour either is rational or norm
guided. One criteria for judging between the approaches, could be explanatory power (scope
multiplied by reliability). If it can be showed empirically that explanations of
unemployment based on norms are better than explanations based on the assumption of
rational behaviour, then "sociological economics" would be the right approach.
Or, conversely, it might turn out that people in general behave rationally, and form
rational beliefs and preferences, thus making economics the king of the social sciences. I
do not know the answer, and I prefer to hedge my bets by examining both rational and
non-rational explanations of the specific phenomenon that I am studying before judging
which I think is most plausible.
As for how to judge plausibility, I agree with Friedman that the ability to predict is
important, but I do not think this is the sole source of plausibility. I also believe
logical considerations should influence my belief in a theory (is it consistent, are the
deductive/inductive inferences correct?. Also information, from experiments and/or
history, can also give some guidance on the plausibility of theories (e.g. testing the
assumption of rational beliefs).
An draft note on sources:
Atkins, Peter (1995), The Periodic Kingdom: A Journey into the Land of Chemical Elements,
London: Phoenix
Becker, Gary S. (199?), The Economic Approach to Human Behaviour, Journal of Political
Economy (?),
Coase, R. H. (1994), Essays on Economics and Economists, Chicago/London: The
University of Chicago Press (Especially ch. 3: Economics and contageous disciplines and
ch. 2 How should economist choose? (arguing against Friedman's methodology))
Elster, Jon. (Most of his papers and books!, Especially: Sour Grapes, "Explaining
Technical Change and Ulysses and the Sirens. See the Jon Elster Page at
www.oocities.org/hmelberg/elster.htm)
Friedman, M. (1953) The methodology of positive economics. In Friedman, M. (ed.), Essays
in Positive Economics. Chicago: University of Chicago Press
Hausman, Daniel M. (1992), The Inexact and Separate Science of Economics,
Cambridge: Cambridge University Press
Landsburg, Steven E. (1993), The Armchair Economist, New York: The Free Press
Mason, Roger S. (1988), The Psychological Economics of Conspicuos Consumption (ch. 10) in
Peter E. Earl (ed.), Psychological Economics, Kluwer Academic Publishers, Boston,
pp. 147-162.
Sen, A. K. (1982), Rational Fools: a critique of the behavioural foundations of economic
theory, in Sen A. K. (ed.), Choice, Welfare and Measurement, Ocford: Blackwell
Solow, Robert M. (1980), On theories of unemployment, American Economic Review,
vol. 70, pp. 1-11.
Further sources (unread so far)
Blaug, M. (1980), The methodology of economics: or How Economists Explain,
Boland, L. A. (1981) On the futility of criticizing the neoclassical maximization
hypothesis. American Economic Review, vol. 71, pp. 1031-6
Dow, A.and Dow, S. C. (1985) Animal Spirits and Rationality. In Lawson, T. and Pesaran, H.
(eds) Keynes' Economics: methodological Issues. Armonk: NY: M. E. Sharpe Inc.
Hamermesh, Daniel S. and Soos, Neal M (1974) An economic theory of suicide, Journal of
Political Economy (January-February)
Hirschman, A. O. (1984), Against Parsimony: three ways of complicating some categories of
economic discourse, American Economic Review, vol. 74 (papers and proceedings), pp.
89-96
Hirschleifer, J. (1985), The expanding domain of economics, American Economic Review,
vol. 75, pp. 53-68
Hounthakker, H. S. (1961), The present state of consumption theory, Econometrica,
vol. 29, pp. 704-40
Marr, W. and Raj, B. (eds.) (1983) How Economists Explain. Lanham: University Press
of America
Rosenberg, A. (1979), Can economics explain everything?, Philosophy of the Social
Sciences, vol. 9, pp. 509-529
Stigler, G. J. (1984), Economics: the imperial science?, Scandinavian Journal of
Economics, vol. 86, pp. 301-313
[Note for bibliographic reference: Melberg, Hans O. (1997), What is economics? A
method or a topic?, www.oocities.org/hmelberg/papers/971018.htm]