Bleg: The scope of abductive reasoning in economics

Posted on by

Time to ask another question.  I was wondering, how much does the form of explanation in economics appear to take the form of abductive reasoning?

Often in economics, we will observe a stylised fact.  We then have a method that can explain that fact in a myriad of ways.  We will then build a model that shows how a given cause will lead to that stylized fact, and pat ourselves on the back.  But if this is really just a form of abductive logic, and there are a myriad of ways to “explain” said stylized fact, how can we have any confidence in our description – how can we really say that we have explained anything?

Posted in Philosophy

About Matt Nolan

Matt Nolan is an economist at Infometrics (although the opinions expressed are independent of the organisation) . Email: matt@tvhe.co.nz; nolan.matt@gmail.com; matt@infometrics.co.nz. Work phone: 04-496-5290

  • http://twitter.com/MarkHubbard33 Mark Hubbard

    How much do you guys get paid again? ;)

    • http://tvhe.co.nz/ Matt Nolan

      The small renumeration we are paid for doing work we love isn’t really relevant to the question of how much economic literature falls under the scope of “abductive reasoning” right.

      I agree we should be paid more for our insights, however it appears the market disagrees – which actually makes the statement I’m agreeing with wrong in this case ;)

  • detmackey

    I’d be looking at how well a model predicted actual change after the fact. But then I haven’t been saying economics isn’t about forecasting and prediction (which is true), and I am still patting myself on the back for a model that predicted a 8.98% change and the actual turned out to be 9.02%. That was two years ago.

    • http://tvhe.co.nz/ Matt Nolan

      There is an inherent conflict between the defence based on prediction (which is often used to denounce realism), and the desire for explanation from models.

      Abductive reasoning seems like an appropriate way of viewing much of the “partial” literature that occurs, at least in micro and industrial economics. To me, we move past this as a discipline through survey articles and meta-analysis – which takes these partial descriptions and weighs them appropriately.

      I’m a big fan of going on and on about whenever an arbitrary number I said is pretty close – it is a measurable achievement, so of course we get something out of it ;)

  • Wasiq Amir

    Your post is very nice and this post having googd information.Thank you for sharing these information