Duration: 1 hours
Seminar Room C, Manor Road Building, Manor Road, Oxford, OX1 3UQ
Economics Dept. workshop series
Consumption is fundamental to macroeconomics, whether in the representative agent Euler equation that drives DSGE models or in the consumption functions of macro-econometric models of the Federal Reserve, HMT-OBR or the BOE’s ‘suite of models’. The rational expectations permanent income model is reviewed in the light of Angus Deaton’s and Chris Carroll’s research. Though the representative agent RE model is wrong, stochastic aggregation can still make aggregate models useful. Better models can clarify the impact of the revolution in credit market architecture, why debt matters and why institutional differences between countries imply major differences in monetary policy transmission. The evidence on linkages between asset prices and consumption is reassessed. The single most important equation in the GE econometric models FRB-US and HMT-OBR is the consumption equation. Both are radically mis-specified, omitting most of the financial accelerator. Joint modelling of household consumption and portfolio decisions, briefly illustrated, is necessary for progress.