How should monetary policy react to sectoral shocks in a world where consumption baskets and demand elasticities vary across households? We present a multi-sector New-Keynesian model with generalized, non-homothetic preferences and inequality. The output gap is governed by a Marginal Consumer Price Index (MCPI), rather than the regular CPI. Policy trade-offs are shaped by two novel wedges in the New-Keynesian Phillips Curve (NKPC). Analytical results and quantitative simulations show that, following a negative shock to necessity sectors, the NKPC is shifted upward, increasing CPI inflation but decreasing the output gap. We find that the optimal policy response is relatively accommodative. 

Empirical evidence points to an increase in the level and dispersion of firm markups starting from around 1980. While there is growing concern that these trends are a result of increased barriers to entry, a more optimistic view maintains that underlying changes in technology can provide an explanation which does not entail welfare losses. To try and disentangle between these competing stories, I use a multi-product model of firm entry, exit and innovation with oligopolistic competition. I calibrate the model for the Manufacturing sector using Compustat data and match selected moments for 1980 and 2010. My findings suggest that the increase in markups can be largely accounted for by a change in the technological process, where product-level productivity improvements have become larger conditional on innovation but happen at a lower rate. With respect to entry barriers, I find that exogenous expansion into new markets has actually gone up and has contributed to an increase in welfare. On the other hand, the fall in the  probability of innovation has reduced welfare gains by around 10pp compared to a counterfactual that holds it unchanged from its initial value.