Job Finding Rate, Separation 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 2005). All the series are quarterly averages of instantaneous transition rates, 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 (2005) and his webpage http://home.uchicago.edu/~shimer/data/flows/.”
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.
All data are quarterly. 1951 is the first quarter of 1951, 1951.25 is the second quarter, etc. These are text files and can be read using any text editor.
Job Finding Rate (1951Q1-2004Q4)
Separation Rate (1951Q1-2004Q4)
Employment-Unemployment Transition Rate (1967Q2-2004Q4)
Employment-Inactivity Transition Rate (1967Q2-2004Q4)
Unemployment-Employment Transition Rate (1967Q2-2004Q4)
Unemployment-Inactivity Transition Rate (1967Q2-2004Q4)
Inactivity-Employment Transition Rate (1967Q2-2004Q4)
Inactivity-Unemployment Transition Rate (1967Q2-2004Q4)
I am also including links to two other useful monthly data series:
Original Data and Source Codes
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.
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 http://www.nber.org/data/cps_basic.html. 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 match_flows.do, 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.
The data collection was supported by the National Science Foundation under Grants 9709881, 0079345, and 0351352. 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.
Last revised: May 26, 2005.
©Robert Shimer, 2005