This dissertation studies the behavior of labor markets over the business cycle. The chapters extend existing literature on the cyclical change in labor markets by considering alternative environments such as on-the-job search and costly screening as well as by estimating a structural labor market search model to understand the effects of different exogenous forces. Chapter 2 uses a standard labor market search model to uncover the cyclical properties of unobserved forcing variables that determine the exogenous state of the aggregate labor market. A structural estimation of the model implies that labor market reallocation as well as implied job separations is strongly procyclical. This result and the recent literature emphasizing the lack of volatility in the standard search models provide the motivation for stressing the role of on-the-job search in creating more volatility for vacancies and unemployment in Chapter 3. A model of on-the-job search with match specific learning is developed in this chapter. Simulations of the model show that job-to-job transitions significantly improve the volatility of vacancies and unemployment. The model implies that firms are more likely to meet employed workers in expansions and that those they meet are more likely to accept firm's job offer because they are more likely to be employed in a low quality match. This introduces strongly procyclical labor market reallocation through procyclical job-to-job transitions. Chapter 4 seeks to determine the quantitative importance of costly screening for the observed excess volatility. A model similar to the one used in Chapter 3 is utilized to emphasize the symmetric incomplete information about match quality. In the model, employers are endowed with a costly screening technology, which enables them to meet with potentially better quality workers. Since this technology is costly, employers decide to use it when aggregate productivity is low, i.e. when the opportunity cost of not producing is low. These countercyclical changes in screening investment improve the cyclical behavior of vacancies. Therefore, this dissertation provides possible answers for the observed high volatility of key labor market aggregates by emphasizing the significance of recruitment behavior and a more through modeling of workers’ search behavior.