Job Finding Rate, Employment Exit Rate, and Gross Worker Flows

The data on this page were constructed as part of the paper “Reassessing the Ins and Outs of Unemployment” (Shimer Review of Economic Dynamics 2012). All the series are quarterly averages of monthly transition probabilities, corrected for time-aggregation. Please see that paper for details and data definitions. Anyone is free to download the data and use it in his/her own research, as long as he/she includes the following statement: “This data was constructed by Robert Shimer. For additional details, please see Shimer (2012).”

If you use the gross worker flow data from before 1976, you must also include the following statement: “The data from June 1967 and December 1975 were tabulated by Joe Ritter and made available by Hoyt Bleakley.” If you know how to get the raw CPS files from this period, I would appreciate it if you let me know.

Main results
All data are quarterly. 1948 is the first quarter of 1948, 1948.25 is the second quarter, etc. These are text files and can be read using any text editor.

Job Finding Probabaility (1948Q1-2007Q1)
Employment Exit Probability (19Q1-2007Q1) (this series was formerly known as the separation probability)
Employment-Unemployment Transition Probability (1967Q2-2007Q2)
Employment-Inactivity Transition Probability (1967Q2-2007Q2)
Unemployment-Employment Transition Probability (1967Q2-2007Q2)
Unemployment-Inactivity Transition Probability (1967Q2-2007Q2)
Inactivity-Employment Transition Probability (1967Q2-2007Q2)
Inactivity-Unemployment Transition Probability (1967Q2-2007Q2)

I am also including links to two other useful monthly data series:
Unemployment Rate (from the BLS, opens in a new window)
Vacancies (from the St. Louis Fed, opens in a new window). Unfortunately, these data are no longer updated.

Original Data and Source Codes

These files are now somewhat outdated and need to be updated to use the most current data.

Job Finding and Employment Exit Probabilities
This file is self-contained. The original data and source codes are compressed in .zip files. Uncompress the files and look at the readme for more instructions. The source code is written in Mathematica. You will need this commercial program to replicate the results; however, the source code should be readable using the free Mathematica viewer, Mathreader.

Gross Worker Flows
To use these files, you must download the basic monthly CPS files from January 1976 to January 2005. The raw data files are available online at The compressed files require about 3GB of disk space. The uncompressed files require about 30GB of space.

Uncompress the raw files, place them in the folder c:\cps, and name them cpsbYYYYMM.raw. Finally, place the uncompressed files you download here in the same directory. Using Stata (version 7.0 or later), type cd "c:\cps", hit enter, and type do match. This will output six text files containing the gross flows data.

Stata will also create many small monthly files. You can delete the files named cpsYYYYMM.dta. You may wish to keep the files named mergYYYYMM.dta. By modifying the file, you can quickly construct measures of gross flows conditional on age, sex, race, and other observable characteristics that are included in the merged files. Please look at the source code and the CPS codebooks for more information on the characteristics that are included in these files.

To correct for time-aggregation, run the Mathematica program three-state.nb. This will output six .dat files containing the instantaneous transition probabilities between employment, unemployment, and inactivity. Those are the six transition rates listed above under Main Results. It will also output .dat files containing the contribution of each of the flows to changes in the unemployment rate and employment-population ratio.

Much of this material is based upon work supported by the National Science Foundation under Grants 9709881, 0079345, and 0351352 and 0648842. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.

©Robert Shimer, 2014-2021