Essays on individual-level wage stickiness and forward guidance

Abstract: Wage stickiness and household heterogeneitySince long, it has been recognized that wage frictions provide realism to, and improve the properties of, macroeconomic models. Recently, another element of realism has generated a large macroeconomic literature: the addition of household heterogeneity, especially through uninsurable idiosyncratic risk on the level of worker productivity. In this chapter, I examine these two elements jointly. I incorporate wage stickiness on the household level into a standard macroeconomic model with household heterogeneity. A standard assumption when wages are sticky is that the labor demand is forcing, i.e., that households commit to supplying the amount of labor that is demanded, even if this is against their will. I show that in this setting, such an assumption is particularly unrealistic, and hence I relax it. I find that in an environment where the households cannot be forced to supply labor, productivity shocks can give rise to spells of severe underemployment -- a proxy for unemployment -- when wages are high relative to the productivity. When wages are low relative to productivity, hours worked rise, but only moderately so.Wage stickiness and household heterogeneity in general equilibriumIn this chapter, I analyze the general-equilibrium implications of the heterogeneous-agents model with sticky wages that I develop in chapter 1. I find that the underemployment risk caused by a household-level wage stickiness has a large impact on the worker’s precautionary motive to save, and hence also on the equilibrium interest rate. Moreover, the wage friction causes a high dispersion of labor supply across households, in turn leading to a low aggregate labor supply, and hence also to a low production. I show that the main findings are robust to variations in the key model parameters.How big is the toolbox of a central banker? Managing expectations with policy-rate forecasts: Evidence from SwedenSome central banks have decided to publish forecasts of their policy rates. Can such forecasts be used to manage market expectations of future policy rates? In this chapter, I use an event study and regression analysis on Swedish high-frequency data to conclude in the affirmative. Surprises in an announced policy-rate forecast by the central bank affect expectations of the future policy rate up to a horizon of approximately a year and a half. However, the response is not one-to-one, but is estimated to be less than one half. It is also decreasing with the forecast horizon. Moreover, I find that the actual decisions of the bank on its current policy rate -- to the extent their choices are surprises -- influence the market expectations. However, this mechanism is only active for short horizons (less than two quarters). The longer-run market expectations on the policy rate are not affected by policy-rate surprises today.