Replication package for: Education and the Margins of Cyclical Adjustment in the Labor Force

1 Replication materials for “Education and the Margins of Cyclical Adjustment in the Labor Force” 1.1 Overview The code in this replication package constructs the analysis files from five data sources U.S. Census Bureau;, The Ohio State University. Columbus, OH: 2025.(n.d.)Bureau of Labor Statistics, U.S. Department of Labor. Produced and distributed by the Center for Human Resource Research (CHRR), The Ohio State University. Columbus, OH: 2025.;Flood, King, Rodgers, Ruggles, Warren, Backman, Chen, Cooper, Richards, Schouweiler and Westberry; U.S. Bureau of Labor Statistics;U.S. Bureau of Economic Analysis using Stata. Code must be run in unix. To build the analysis datasets execute “code/build.do”. After, to run the analysis execute analysis.do. Note, analysis of the SIPP data takes 3-4 days due to long computation time in estimating the standard errors. 1.2 Data Availability and Provenance Statements Survey of Income and Program Participation (SIPP): The SIPP is a nationally representative longitudinal survey that provides comprehensive information on the dynamics of income and employment. SIPP interviews individuals for several years and provides monthly data about changes in household and family composition and economic circumstances over time. SIPP is a household-based survey designed as a continuous series of national panels. Each panel generally features a large sample of households that are interviewed multiple times over a multi-year period. SIPP data are public use and can be freely obtained from the U.S. Census Bureau. National Longitudinal Survey of Youth, 1979 cohort (NLSY97): The NLSY97 consists of a nationally representative sample of 8,984 men and women born during the years 1980 through 1984 and living in the United States at the time of the initial survey in 1997. Participants were ages 12 to 16 as of December 31, 1996. Interviews were conducted annually from 1997 to 2011 and biennially since then. The ongoing cohort has been surveyed 21 times as of date. Data are available from Round 1 (1997–98) through Round 20 (2021–22). The NLSY97 collects extensive information on respondents’ labor market behavior and educational experiences. The NLSY79 survey is sponsored and directed by the U.S. Bureau of Labor Statistics and managed by the Center for Human Resource Research (CHRR) at The Ohio State University. Interviews are conducted by the National Opinion Research Center (NORC) at the University of Chicago. NLS public-use data for each cohort are available at no cost from the U.S. Bureau of Economic Analysis via Investigator, an online search and extraction site that enables you to review NLS variables and create your own data sets. It is not necessary to get an account to browse data, but an account is necessary to save datasets online. The Investigator User’s Guide describes how to use this website. Current Population Survey (CPS): The CPS is sponsored jointly by the U.S. Census 1 Bureau and the U.S. Bureau of Labor Statistics, is the primary source of labor force statistics for the population of the United States. IPUMS CPS harmonizes microdata from the monthly U.S. survey and topical supplements. The monthly survey includes demographic information, and employment data. The monthly survey is conducted as a rotating panel in which respondents are surveyed for four consecutive months in two consecutive years. The “earner study” topical supplement includes earnings data and the “job tenure and occupational mobility supplement” includes data on tenure in the current job. The job tenure supplement was conducted in January 1983, 87, 91, 2002 and biyearly thereafter as well as biyearly in February from 1996 to 2000. This paper makes use of job tenure supplement observations linked to wage and employment data from the monthly surveys and earner study via the rotating panel structure. Researchers can access the IPUMS CPS data without charge by completing this registration form. Unemployment Rate (UNRATE): The unemployment rate represents the number of unemployed as a percentage of the labor force. Labor force data are restricted to people 16 years of age and older, who currently reside in 1 of the 50 states or the District of Columbia, who do not reside in institutions (e.g., penal and mental facilities, homes for the aged), and who are not on active duty in the Armed Forces. This rate is also defined as the U-3 measure of labor underutilization. The series comes from the “Current Population Survey (Household Survey)” and can be downloaded freely from the Federal Reserve Bank of St. Louis’s FRED. Implicit Price Deflator: The gross domestic product implicit price deflator measures changes in the prices of goods and services produced in the United States, including those exported to other countries. Prices of imports are excluded. The U.S. Bureau of Economic Analysis and can be downloaded freely from the Federal Reserve Bank of St. Louis’s FRED. 1.2.1 Statement about Rights ■I certify that the author(s) of the manuscript have legitimate access to and permission to use the data used in this manuscript. 1.2.2 Summary of Availability ■All data are publicly available. 1.2.3 Details on each Data Source 1.2.4 Survey of Income and Program Participation The paper uses SIPP data (U.S. Census Bureau,n.d.). The following program is executed by code/build.do and downloads the raw SIPP data to the appropriate directories in rawdata/SIPP. Acquisition: rawdata/SIPP/raw/download data cd.sh 2 1.2.5 National Logitudinal Survey of Youth, 79 The paper uses NSLY79 data (Bureau of Labor Statistics, U.S. Department of Labor. Produced and distributed by the Center for Human Resource Research , CHRR). Datafiles and codebooks: rawdata/NLSY/build/input/ •C SAMPWEIGHT.dct •C SAMPWEIGHT-value-labels.do •EMPLOYERS ALL UNION.dct •EMPLOYERS ALL UNION-value-labels.do •EMPLOYERS ALL WHYLEFT.dct •EMPLOYERS ALL WHYLEFT-value-labels.do •DATE OF BIRTH.dct •DATE OF BIRTH-value-labels.do •EMPLOYERS ALL COW.dct •EMPLOYERS ALL COW-value-labels.do •HGCREV.dct •HGCREV-value-labels.do •EMPLOYERS ALL HOURSWEEK.dct •EMPLOYERS ALL HOURSWEEK-value-labels.do •JOB WK NUM DUAL JOB1.dct •JOB WK NUM DUAL JOB1-value-labels.do •EMPLOYERS ALL HRLY WAGE.dct •EMPLOYERS ALL HRLY WAGE-value-labels.do •JOB WK NUM DUAL JOB2.dct •JOB WK NUM DUAL JOB2-value-labels.do •EMPLOYERS ALL IND.dct •EMPLOYERS ALL IND-value-labels.do 3 •JOB WK NUM DUAL JOB3.dct •JOB WK NUM DUAL JOB3-value-labels.do •EMPLOYERS ALL NUM ARRAY.dct •EMPLOYERS ALL NUM ARRAY-value-labels.do •JOB WK NUM DUAL JOB4.dct •JOB WK NUM DUAL JOB4-value-labels.do •EMPLOYERS ALL STADATE.dct •EMPLOYERS ALL STADATE-value-labels.do •MARSTAT-KEY.dct •MARSTAT-KEY-value-labels.do •EMPLOYERS ALL STOPDATE.dct •EMPLOYERS ALL STOPDATE-value-labels.do •EMPLOYERS ALL TENURE.dct •EMPLOYERS ALL TENURE-value-labels.do •RNI.dct •RNI-value-labels.do •EMPLOYERS ALL TIMERATE.dct •EMPLOYERS ALL TIMERATE-value-labels.do •STATUS WK NUM.dct •STATUS WK NUM-value-labels.do •rounds.dta 4 1.2.6 Current Population Survey The paper uses IPUMS CPS data (Flood et al.,2023). CPS source data provided by the United States Census Bureau and Bureau of Labor Statistics are the underling data used by IPUMS CPS. The IPUMS CPS extract required is large and processed data are provided. A researcher who wishes to recreate the processed data from a raw extract can submit a extract request to IPUMS CPS for the variables and datasets listed in the provided rawdata/CPS/cps 00120.cbk (which can be read using any text editor). The corresponding .do,.dat.gz, and .do. These can be processed into cps ‘vintage’ impute.dta.gz using the code commented out under CPS in code/build.do. The local ‘vintage’ will need to be updated throughout to corresponding to the vintage of the extract obtained by the replicator. Run time may be as long as 24 hours and employs parallel processing. Datafile: rawdata/CPS/cps 00120 impute.dta.gz 1.2.7 Unemployment Rate (UNRATE) and Implicit Price Deflator Datafiles: data/FRED/ipd.dta, data/FRED/uerate 2025.dta 1.3 Computational requirements 1.3.1 Software Requirements ■The replication package contains one or more programs to install all dependencies and set up the necessary directory structure. Stata (code was last run with version 18) Portions of the code use bash scripting, which may require Linux. 1.3.2 Controlled Randomness ■No Pseudo random generator is used in the analysis described here. 1.3.3 Memory, Runtime, Storage Requirements The code was last run on a linux cluster using 6 cores and 100g of memory.. 1.4 Description of programs/code •Replication materials are organized into subdirectories code, data, figures, rawdata. Within each of code, data, and rawdata there are subsubdirectories for each data source: SIPP, NLSY, CPS, FRED. Code must be run in unix. To build the analysis datasets execute ”code/build.do”. After, to run the analysis execute analysis.do. Note, analysis of the SIPP data takes 3-4 days due to long computation time in estimating the standard errors. 5 1.5 Instructions to Replicators •Edit code/build.do and code/analysis.do to adjust the default paths •Run code/build.do to set up the working environment. •Run code/analysis.do to run all analyses. 1.6 List of tables and programs The provided code reproduces: ■All numbers provided in text in the paper with the exception of Table A2, wherein the citations are provided. ■All tables and figures in the paper 1.7 References References Bureau of Labor Statistics, U.S. Department of Labor. Produced and distributed by the Center for Human Resource Research (CHRR), The Ohio State University. Columbus, OH: 2025., “National Longitudinal Survey of Youth 1979 cohort,” https://nlsinfo.org. Flood, Sarah, Miriam King, Renae Rodgers, Steven Ruggles, J. Robert Warren, Daniel Backman, Annie Chen, Grace Cooper, Stephanie Richards, Megan Schouweiler, and Michael Westberry, “Integrated Public Use Microdata Series, Current Population Survey,” 2023. U.S. Bureau of Economic Analysis, “Gross Domestic Product: Implicit Price Deflator [GDPDEF],” retrieved from FRED, Federal Reserve Bank of St. Louis; https://fred.stlouisfed.org/series/GDPDEF. U.S. Bureau of Labor Statistics, “Unemployment Rate, UNRATE,” retrieved from FRED, Federal Reserve Bank of St. Louis; https://fred.stlouisfed.org/series/UNRATE. U.S. Census Bureau, “Survey of Income and Program Participation,” http://www. census.gov/programs-surveys/sipp.html. 6 Table 1: List of tables and programs Figure/Table Programs (in code/) Output files (in figures/) Table 1 NLSY NLSY/NLSY setup.do Table1 B1 NLSY.txt Table 1 SIPP SIPP/ReStud2 UserCost SIPP inst exp robust p.do Tables1 2 A1 A5 B1.txt Table 1 CPS CPS/RESTUD UserCost CPS.do Tables1 B1 CPS2024.txt Figure 1 NLSY NLSY/NLSY setup.do Figure1 NLSY.pdf Figure 1 SIPP SIPP/ReStud2 UserCost SIPP inst exp robust p.do Figure1 SIPP.pdf Figure 1 CPS CPS/RESTUD UserCost CPS.do Figure1 CPS 2024.pdf Table 2 NLSY NLSY/NLSY byEd p.do Table2 NLSY.txt Table 2 SIPP SIPP/ReStud2 UserCost SIPP inst exp robust p.do Tables1 2 A1 A5 B1.txt Figure 2 NLSY NLSY/NLSY byEd p.do Figure2 NLSY.pdf Figure 2 SIPP SIPP/ReStud2 UserCost SIPP inst exp robust p.do Figure2 2.pdf Table 3 SIPP/ReStud2 UserCost SIPP inst exp decomp x2 p.do Tables3.txt Figure 3 SIPP/ReStud2 UserCost SIPP inst exp age p.do Figure3.pdf Table 4 SIPP/ReStud2 UserCost SIPP inst exp age marg p.do Tables4 5.txt Table 5 SIPP/ReStud2 UserCost SIPP inst exp age marg p.do Tables4 5.txt Figure 4 SIPP/ReStud2 UserCost SIPP inst exp asymetric linten p.do Figure5.txt Table A1 SIPP/ReStud2 UserCost SIPP inst exp robust p.do Tables1 2 A1 A5 B1.txt Table A2 Citations provided Table A3 NLSY/NLSY 1v2 step.do TableA3.txt Footnote 30 NLSY/NLSY 1v2 step.do Footnote30.txt Table A4 NLSY/NLSY cumtight R2 p.do Table A5.txt Table A5 SIPP/ReStud2 UserCost SIPP inst exp robust p.do Tables1 2 A1 A5 B1.txt Table B1 NLSY NLSY/NLSY setup.do Figure1 NLSY.pdf Table B1 SIPP SIPP/ReStud2 UserCost SIPP inst exp robust p.do Tables1 2 A1 A5 B1.txt Table B1 CPS CPS/RESTUD UserCost CPS.do Tables1 B1 CPS2024.txt Table C1 CPS/RESTUD UserCost CPS.do Tables1 B1 CPS2024.txt Figure C1 CPS/RESTUD UserCost CPS.do FigureC1 CPS2024.txt Table D2 NLSY NLSY/NLSY byEd pt.do TableD2 NLSY.txt Table D2 SIPP SIPP/ReStud2 UserCost SIPP inst exp robust pt.do TableD2.txt Figure D2 NLSY NLSY/NLSY byEd pt.do FigureD2 NLSY.txt Figure D2 SIPP SIPP/ReStud2 UserCost SIPP inst exp robust pt.do FigureD2.pdf Table D3 SIPP/ReStud2 UserCost SIPP inst exp decomp x2 pt.do TablesD3.txt Figure D3 SIPP/ReStud2 UserCost SIPP inst exp age pt.do FigureD3.pdf Table D4 SIPP/ReStud2 UserCost SIPP inst exp age marg pt.do TablesD4 D5.txt Table D5 SIPP/ReStud2 UserCost SIPP inst exp age marg pt.do TablesD4 D5.txt Figure D4 SIPP/ReStud2 UserCost SIPP inst exp asymetric linten pt.do FigureD5.txt Table D4 SIPP/ReStud2 UserCost SIPP inst exp asymetric linten pt.do FigureD5.txt Table D.A4 NLSY/NLSY cumtight R2 pt.do Table DA4.txt Table D.A5 SIPP/ReStud2 UserCost SIPP inst exp robust pt.do TablesD2 DA5.txt 7