Downward nominal wage rigidity, money illusion, and irreversibility
This thesis seeks to make three related contributions to our understanding of the causes and implications of downward nominal wage rigidity, the nature of money illusion on behalf of workers, and the theoretical treatment of irreversibility in factor demand and wage setting. Chapter 1 seeks to contribute to the literature on downward nominal wage rigidity (DNWR) along two dimensions. First, I formulate and solve an explicit model of wage- setting in the presence of worker resistance to nominal wage cuts - something that has previously been considered intractable. In particular, I show that this resistance renders wage increases (partially) irreversible. Second, using this model, one can explain why previous estimates of the macroeconomic effects of DNWR have been so weak despite remarkably robust microeconomic evidence. In particular, one can show that previous studies have neglected the possibility that DNWR can lead to a compression of wage increases as well as decreases. Thus, the literature may have been overstating the costs of DNWR to firms. Using micro-data for the US and Great Britain, I find robust evidence in support of the predictions of the model. In the light of this evidence, Chapter 1 concludes that increased wage pressure due to DNWR may not be as large as previously envisaged, but that the behavioural implications of DNWR in respect of the reaction of workers to nominal wage cuts remain significant. Chapter 2 then contrasts the implications of two proposed models of downward nominal wage rigidity - those based on the form of market contracts (MacLeod and Malcomson ; Holden ), and that based on money illusion explored in Chapter 1. In particular, I identify a method of distinguishing between these two foundations empirically, by observing how the distribution of wage changes varies with the rate of inflation. I find evidence that at least part of the observed rigidity cannot be easily explained by contract models, but can be explained in the context of a model with money illusion. Finally, Chapter 3 extends some of the theoretical developments of Chapter 1 with respect to models of irreversibility. In particular, Chapter 3 presents analytical results for models of dynamic factor demand in the presence of irreversibility in discrete time. It builds on previous work on irreversibility in the investment (Dixit and Pindyck ) and labour demand (Bentolila and Bertola ) literatures which use a continuous time. Brownian framework. I show that, whilst there are parallels between the discrete time models and their continuous time counterparts, the analytics in discrete time allow a more general treatment, principally by allowing the relaxation of the assumption of shocks following a unit root. I then explore the effects of relaxing this assumption on optimal factor demand.