This dissertation contains two themes. The first involves the relationship between economic uncertainty and employment/unemployment determination, the second involves the institution of firing costs, in particular its effect on employment over the business cycle. Chapter 2 introduces uncertainty about labour productivity into the dynamic turnover-training model of Hoon and Phelps (1992). This makes hiring an investment under uncertainty. I get an explicit solution for the Bellman equation for the representative firm's optimization problem and find the optimal path for employment at the industry level. It is shown that an increase in the level of uncertainty reduces the shadow price of employees, the hire rate and optimal wage, and shifts the path for industry employment to a lower level. Chapter 3 then moves on to test the hypothesis using both data for all the OECD countries and also panel data for British industries. I find that those industries which have experienced the most volatility - measured by their average share prices - also have had the slowest employment growth, other things equal. Similarly, epochs of high uncertainty, measured by inflation and exchange rate movements, have accompanied low job creation in the OECD countries. The policy implication is that predictable economic policies and stability may reduce the equilibrium rate of unemployment. Chapter 4 looks at the determinants of the effectiveness of the institution of statutory firing costs. In the past, measures of firing restrictions have only included the size of the redundancy payments. I show that this is seriously inadequate. Thus the effectiveness of firing restrictions depends on the persistence of shocks to labour demand, the rate of interest, the rate at which workers quit, the degree of uncertainty about future productivity and economic growth. The implication is that a given level of redundancy payments may either have a very large or a very small effect on the number of redundancies, depending on the economic environment.