The objectives of this paper are : first, to quantify the stabilization welfare gains from commitment; second, to examine how commitment to an optimal rule can be sustained as an equilibrium and third, to find a simple interest rate rule that closely approximates the optimal commitment one. We utilize an influential empirical micro-founded DSGE model, the euro area model of Smets and Wouters (2003), and a quadratic approximation of the representative household's utility as the welfare criterion. Importantly, we impose the effect of a nominal interest rate zero lower bound. In contrast with previous studies, we find significant stabilization gains from commitment: our central estimate is a 0.4 − 0.5% equivalent permanent increase in consumption, but in a variant with a higher degree of price stickiness, gains of over 2% are found. We also find that a simple optimized commitment rule with the nominal interest rate responding to current inflation and the real wage closely mimics the optimal rule.
JEL Classification: E52, E37, E58Keywords: Monetary rules, commitment, discretion, welfare gains.
ECB Working Paper Series No 709 January 2007
Non-technical SummaryThis paper has three principle objectives. First, to quantify the stabilization gains from commitment in terms of household welfare. Second, to examine how commitment to an optimal or approximately optimal rule can be sustained as an equilibrium in which reneging hardly ever occurs. And finally, to find a simple interest rate rule that closely approximates the optimal commitment (and complex) rule.We utilize an influential empirical micro-founded dynamic stochastic general equilibrium (DSGE) model of the euro area in which there are four sources of time-inconsistency: from forward-looking pricing, consumption, investment and wage setting. In the absence of commitment, following a shock which diverts the economy from its steady state and given expectations of inflation, the opportunist policy-maker can increase or decrease output by reducing or increasing the interest rate which increases or decreases inflation. This results in a higher variability of inflation and the nominal interest rate under discretion. The latter means that the interest rate zero lower bound constraint is tighter under discretion and its presence increases the stabilization gains from commitment. The constraint can be relaxed by increasing the steady state inflation rate, but at a cost of an increase in the deterministic component of the welfare loss.In terms of methodology our welfare-based loss function uses the 'small distortions' quadratic approximation to the consumer's utility which is accurate if the steady state is close to the social optimum. In assessing this condition we highlight a neglected aspect of typical New Keynesian models: external habit in consumption tends to make labour supply and the natural rate of output too high compared with the social optimum. If the habit effect is sufficiently high and labour market and product market distortions are not too big then, ...