Question 1: Are economic theories and models developed using induction or deduction?

My view: If we gather historical evidence and figure "so, people tend to be like this and that so let's call this the Law of somethingsomething", then the answer is clearly induction.

Question 2: Are economic theories judged using verificationism or falsificationism?

My view: It seems to be verificationism: if the theories and models seem to be at least somewhat in accordance with reality, then it is considered a verification of the theory.


As regards the induction/deduction division, economics uses both, and they usually start by some induction followed by mathematizing it and then using deduction.

In the older days, economists were based more on their own personal experience and observation as to how people behave economically. Just browse any classic economics book from the 18th, 19th and even pre-war 20th century and this becomes obvious. They then embarked on lengthy verbal arguments to connect these instances of experience (in the modern era this approach has partly survived, but formalized into the "representative agent" analytical framework).

In modern days, both stages have changed: At least for fundamental advances, what can be detected as long-term trend is that solid empirical evidence (being statistical or other), that can claim some generality, has become a required springboard for induction and a new theoretical economic model. Just one example of a seminal such paper,

Aiyagari, S.R. (1994). Uninsured idiosyncratic risk and aggregate saving. The Quarterly Journal of Economics, 109(3): pp.659-684.

is a paper where the author packs considerable statistical evidence and theoretical and empirical references in two pages before presenting his model of (what the paper title says), as a way of validating his approach.

The second stage has become mainly mathematical rather than verbal: one lets its model roll and see what results it will produce (this is the deduction step). This synthetic approach is necessitated because social phenomena do not have the repeatability observed in the natural world (although I would dare to say that there are many aspects of economic behavior that have remained virtually unchanged for at least all written history, and around the globe).

Of course since nowadays science is produced in an industrialized way, a million economics papers out there are purely works on existing theoretical frameworks, so this strand is pure deduction.

As regards the verificationism/falsificationism contrast, it appears that the OP uses the words in a rather relaxed manner, away from their stern meaning and the philosophical schools behind them. In that sense then, yes, economics looks for "adequate representations of reality", i.e. it sides with "verificationism": an inside joke is that "no economic theory was ever abandoned because it did not conform with the data"...

For such matters, Mark Blaug's inside polemic The methodology of economics: Or, how economists explain has not lost its bite.

The helpful comment below suggested also looking at abductive inference for how economists go methodologically about understanding the world.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.