Florian Guibelin*, Etienne Vaccaro-Grange**
Anushka Chawla : anushka.chawla[at]univ-amu.fr
Laura Sénécal : laura.senecal[at]univ-amu.fr
Carolina Ulloa Suarez : carolina.ulloa-suarez[at]univ-amu.fr
*I use a matching model where working time and wage are bargained between heterogeneous firms and workers. I compare the effect of three mechanisms on labor market variables and utilities : (1) a pure Unemployment Insurance system, benefits are proportional to wages; (2) a Basic Income system, everyone receives the same allocation, no matter his situation; (3) a Hybrid system, everyone receives the basic income, except for unemployed individuals for whom unemployment benefits are greater than the universal allocation. If the income tax rate is the same in the three systems and the basic income high enough, working time, net income, labor demand and workers' utility are greater in BI scheme than in UB mechanism. However, high-skilled unemployed people are worse off in the former. A hybrid system is better for low-skilled individuals than the pure UB scheme, but not necessarily for high-skilled workers.
**The structural relationship between price inflation and measures of real economic activity - the Phillips curve – is said to be dead. Using a canonical New Keynesian model, we explain why existing estimates of its slope are biased, and that the apparent flattening of the Phillips curve is due to omitted supply side factors. We then proceed in two steps: in step one, we estimate a structural VAR model with sign restrictions to purge the unconditional data for supply shocks. Step two is to estimate Phillips curve regressions using the purged data. Given this restriction, the main results can be summarised as follows: first, in line with previous literature, the Phillips curve is apparently dead when we estimate it based on unconditional data. And it has been dead since the beginning of the Great Moderation. Second, once we condition on the data filtered for supply shocks, the Phillips curve comes back alive and well: the slope of the Phillips curve has remained relatively constant in the last decades. If anything, it has increased slightly in recent years.