2
$\begingroup$

Since i had my "Introduction to advanced macroeconomics" and discovered DSGE i've been wondering if IS-LM is used in research, i know it was used in the 80's or so, but now?

$\endgroup$
1
$\begingroup$

Yes, it is still used a recent example of IS-LM model used in literature is (Bonyah et al. 2019).

Also, it is important to understand those two are not substitutes for each other. IS-LM is a general equilibrium (GE) model. The way how you get dynamic stochastic general equilibrium (DSGE) model is that you take a general equilibrium model and add dynamism and uncertainty there.

As a student you were probably exposed more to the DSGE model based on real business cycle (RBC) model (which is based on Neoclassical GE models) as they are simpler, but New Keynesian DSGE models which are in essence DSGE version of IS-LM exists as well (see discussion of New Keynesian DSGE in Frische & Größl, 2010) and are relatively widely used (in fact I would say that after 2008 they became increasingly popular in the literature).

You will see DSGE models dominating over GE models in the literature, but this is because GE models are mostly used to derive some theoretical insights into how economy works, whereas DSGE can be used for actual practical simulations or even forecasting. You can't make any serious forecasts with theoretical GE model - no matter whether it is IS-LM GE model or some other one (only predictions about variables and relationship but that should not be confused with forecasting). Consequently, you will see DSGE models dominating because it makes sense to make separate paper about DSGE model calibrated for let's say economy of the Netherlands as in Lafourcade and de Wind (2012) (coincidentally they actually apply IS-LM based DSGE), whereas applying any GE whether IS-LM or some Neoclassical one to a particular country would not be very useful (at least not much beyond what you can learn from the general version already). So for every one theoretical GE model, no matter what variety, you can easily find more than dozen DSGE models.

$\endgroup$
9
  • $\begingroup$ One needs a very permissive notion of general equilibrium models to count IS-LM among it. Also, if you model the same commodity at different times and states of the world as different commodities, the standard Arrow-Debreu model can model time and uncertainty. The IS-LM model plays no serious role in current research, and the cited paper is not from an economics journal. $\endgroup$ Mar 17 at 17:47
  • $\begingroup$ @MichaelGreinecker care to elaborate on the first sentence? That might apply maybe to canonical versions but I think that would be too harsh for modern variations. Also, you are right that it does not play big role in current research but that is primarily because most of modern macroeconomics is empirical. In theoretical research it still plays a role, you can just have look at Woodford Interest & Prices where that is one of the workhorse models and papers post 2000 using it are cited there. Also note, modern versions have rigorous microfoundations and can be viewed as special cases of $\endgroup$
    – 1muflon1
    Mar 17 at 18:17
  • $\begingroup$ walrasian model. $\endgroup$
    – 1muflon1
    Mar 17 at 18:17
  • 1
    $\begingroup$ Sure, I hope my comment did not come off as too negative. I simply interpreted OPs intend differently, and there is not much context given. $\endgroup$ Mar 17 at 18:45
  • 1
    $\begingroup$ @MichaelGreinecker no worries I took in good faith and in the NL where I am such direct comments are part of academia so I am used to that $\endgroup$
    – 1muflon1
    Mar 17 at 18:50

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.