Demand shocks, new keynesian model and supply effects of monetary policy

Demand shocks likely play a key role in driving business cycles. However, in the standard new keynesian model, the monetary policy reaction to these shocks have a supply side effect. The change in real rate affects the marginal utility of consumption generating an income effect on labor supply. Wages, inflation and through monetary policy, aggregate demand will increase. This supply side effect have a surprising importance for the model, especially when the sensi- tivity of aggregate demand to interest rate is low. A demand shock will have a large impact (close to one) on output, but a very small one on the output gap. The limited monetary policy movement induced by the taylor rule remains very close to the natural rate of interest. There are nearly no differences between the sticky price and the flexible price model. It represents a very disappointing result, the entire purpose of sticky prices being to generate inefficiencies when the aggregate demand is hit. Coupled with very tiny empirical support for this supply side effect of monetary policy, it suggests to explore the theoretical possibilities to kill this ef- fect. First, we review the two ways the literature have proposed, nonseparable preferences and sticky wages. The main drawback is a strong reliance on very specific assumption for the labor market. We explore an alternative approach. We attempt a radical departure from traditionnal assumption about the optimizing behavior of the representative agent. Instead of optimizing simulatneously with respect to hours, consumption and saving, the household decomposes the problem in two steps. First, the agent chooses between labor income and hours. Second, he optimizes between consumption and saving. The interest is to disentangle the income effect which affects the labor equation and those affecting the intertemporal choice. Thus it is possible to reduce the wealth effect on labor supply whereas keeping a low sensitivity of consumption to interest rate. This flexible approach also allows to challenge the effect of interest rate on wealth offering a potential explanation for small effects of interest rate on both labor supply and consumption whereas keeping large income effects.