On Identification Issues in Business Cycle Accounting Models

Since its introduction by Chari et al. (2018), Business Cycle Accounting (BCA) exercises have become widespread. Much attention has been devoted to the results of such exercises and to methodological departures from the baseline methodology. Little attention has been paid to identification issues within these classes of models, despite the methodology typically involving estimating relatively large scale dynamic stochastic general equilibrium models. In this paper we investigate whether such issues are of concern in the original methodology and in an extension proposed by Sustek (2011) called Monetary BCA. We resort to two types of identification tests in population. One concerns strict identification as theorized by Komuner and Ng (2011), while the other deals both with strict and weak identification as in Iskrev (2015). As to the former, when restricting the estimation to the parameters governing the latent variable's laws of motion, we find that both in the BCA and MBCA framework, all parameters fulfill the requirements for strict identification. If instead we estimate all structural parameters of the model jointly, both frameworks show strict identification failures in several parameters. These results hold for both tests. We show that restricting estimation of some deep parameters can obviate such failures. When we explore weak identification issues, we find that they affect both models. They arise from the fact that many of the estimated parameters do not have a distinct effect on the likelihood. Most importantly, we explore the extent to which these weak identification problems affect the main economic takeaways and find that the identification deficiencies are not relevant for the standard BCA model. Finally, we compute some statistics of interest to practitioners of the BCA methodology.