Date

2013

Author

Ulucan, Hakan

Metadata

Show full item record

Item Usage Stats

104
views

44
downloads

Cite This

The main objective of this dissertation is to investigate the sources of unemployment fluctuations. Our main purpose is to assess the validity of debated conclusions of Shimer (2012) from a new perspective and re-analyze/re-harmonize his research question using alternative methods in five steps. First, we conduct a preliminary empirical exercise by replicating Baker (1992) with new data and find that Shimer’s conclusion that “the movements in the exit rate from unemployment are the main determinant of unemployment fluctuations” is not that obvious. Second, we apply Shimer’s method to demographic subgroups to see if his conclusions change at the group-level or not. We find that there is considerable heterogeneity across sub-groups in terms of the explanatory powers of the job finding probability versus the exit probability. Thirdly, we extended Shimer’s model by incorporating a new labor market state, “self employment,” to see if his conclusions change. We find that his results are mostly unaltered. But, we document important facts regarding the cyclical properties of the transition to and from self-employment. Fourth, we ask if his results are due to ignored measurement errors. We document that measurement errors have some role, but correcting for them does not alter Shimer’s conclusions substantially. Finally, we take a closer look at the mechanics of Shimer’s statistical model. We observe that focusing on rates instead of numbers may be underemphasizing fluctuations in entry and overemphasizing those in exit from unemployment.

Subject Keywords

Unemployment., Unemployed., Precarious employment., Self-employed., Part-time self-employment.

URI

http://etd.lib.metu.edu.tr/upload/12616528/index.pdf
https://hdl.handle.net/11511/22895

Collections

Graduate School of Social Sciences, Thesis