Checking relevant frictions

In Chari’s business cycle accounting model, you can use wedges to identify which class of frictions are relevant for a particular episode of the business cycle. In Smet and Wouters, they include a lot of frictions and check one at a time which frictions are relevant for the business cycle. Do these two approaches converge? Like if the wage stickiness parameter contributes strongly to the business cycle in Smets and Wouters, then the labor wedge should contribute strongly to the business cycle in Chari’s BCA model, using the same data?

Yes and no. To be consistent with the data, frictions need to map into wedges. But the wedges do not have cross-equation restrictions. Therefore, a friction may map into several wedges at ones.

Oh I see. Many thanks!! Perhaps, one should first use a bca model to identify a class of frictions, and then check their contribution to the business cycle as Smets and Wouters do.

May I ask a related question? You can’t use a set of frictions in a model just because it is ‘standard’, right? I mean some papers sometimes say, we introduce habit formation, investment adjustment and so on…because these frictions are standard. Maybe that is true for economies with a lot of dsge papers…but not for economies with few dsge papers, I guess.

The usual way would be to include these frictions as general possibility and then estimate their strength based on the data. After all, we know that these frictions are necessary to explain the data for pretty much all economies.