Dear MPC Members, Aug 6, 2019, 9:35 AM

I came across the attached article recently,

Srnicek, N. AI & Soc (2018) Eyes of the State: How Central Banks Think

which provides a brilliant analysis of decision making processes at the central bank (in general). There are many peripheral matters of interest, such as the sociology of decision making process, interplay between politics and finance, but of central importance is the use of models.

Models can be used to outsource thinking (to some extent). They systematically highlight some factors and obscure others. The complaint it was systematic weaknesses of the DSGE models that led to the Global Financial Crisis receives a balanced analysis.

After reading this, I have come to the conclusion that we should invest serious time in understanding the inner workings of current models at State Bank, because they are based on RBC and DSGE - updated in many ways to third generation - but still sharing their fundamental weaknesses.

Most important in my view is the fact that these models were developed in USA and UK and are used in economies which are fundamentally different from ours in many ways. These differences are not reflected in the models we use, since we just borrow them without making necessary adaptations. For instance, everyone knows about the huge informal sector, but this is not part of the model - There are many other ways in which our economy is significantly different from USA. By outsourcing our thinking to a model which starts by assuming that our economy is sufficiently similar to the USA that the same modelling principles apply, I think that we are throwing out the baby with the bathwater.