Please cite as
Koren, Miklós, and Rita Pető. 2020. “Business Disruptions from Social Distancing.” arXiv. http://arxiv.org/abs/2003.13983.
@ARTICLE{Koren2020-mk, title = "Business disruptions from social distancing", author = "Koren, Mikl{\'o}s and Pet{\H o}, Rita", journal = "Covid Economics", number = 2, institution = "Centre for Economic Policy Research", year = 2020, archivePrefix = "arXiv", primaryClass = "econ.GN", eprint = "2003.13983" }
If you are interested in replicating the version we submitted to PLOS ONE, please use Release version 1.1
The version published on arXiv removes some occupations where telecommunication can more easily be used. Code and data are available at Release version 1.2
Please cite this version as
Koren, Miklós, and Rita Pető. 2020. “Business Disruptions from Social Distancing.” arXiv [econ.GN]. arXiv. http://arxiv.org/abs/2003.13983.
The version compares our occupation indexes to that of Dingel and Neiman (2020) and cite several pieces of parallel research using O* NET data. Code and data are available at Release version 1.3
All text (*.md, *.txt, *.tex, *.pdf) are CC-BY-4.0. All code (*.do, Makefile) are subject to the 3-clause BSD license. All derived data (data/derived/*) are subject to Open Database License. Please respect to copyright and license terms of original data vendors (data/raw/*).
All the data used in the analysis, including raw data, is available for reuse with permissive licenses. Raw data are saved in the folder data/raw/. The Makefile in each folder shows the URLs used to download the data.
U.S. Department of Labor/Employment and Training Administration, 2020. "O* NET Online." Downloaded 2020-03-12.
CC-BY-4.0 https://www.onetonline.org/help/license
U.S. Bureau of Labor Statistics. 2020. "Current Employment Statistics." https://www.bls.gov/ces/ Downloaded 2020-03-15.
Public domain: https://www.bls.gov/bls/linksite.htm
U.S. Bureau of Labor Statistics. 2018. "National Employment Matrix." https://www.bls.gov/emp/data/occupational-data.htm Downloaded 2020-03-15.
Public domain: https://www.bls.gov/bls/linksite.htm
U.S. Bureau of Labor Statistics. 2019. "O* NET-SOC to Occupational Outlook Handbook Crosswalk." https://www.bls.gov/emp/classifications-crosswalks/nem-onet-to-soc-crosswalk.xlsx Downloaded 2020-03-15.
Public domain: https://www.bls.gov/bls/linksite.htm
Wick, Marc, Christophe Boutreux et al. 2020. "GeoNames geographical database." Available at: https://www.geonames.org/ Accessed on 2020-03-22.
U.S. Bureau of the Census. 2017. "County Business Patterns." Available at https://www.census.gov/programs-surveys/cbp.html
https://www.census.gov/data/developers/about/terms-of-service.html
Analysis uses Stata 15.1 without any additional Stata packages.
The entry point for analysis is analysis/Makefile, which can be run by GNU Make on any Unix-like system by
cd analysis
make
The dependence of outputs on code and input data is captured in the respective Makefiles.
We have used Mac OS X, but all the code should run on Linux and Windows platforms, too.
The analysis takes a few minutes on a standard laptop.
- Raw data are in
data/raw/<vendor>/<dataset>. This data is saved as it has been received from the data publisher, downloaded by the respective Makefiles. Each folder has aREADME.mdwith data citation and license terms. - Clean data are in
data/clean/<dataset>. Each folder has aMakefilethat specifies the steps of data cleaning. - Analysis data are in
data/derived/<topic>. Each folder has aMakefilethat specifies the steps of sample specification, variable selection and modification. - Analysis codes are in
analysis/. AMakefilespecifies how the .do files have to be run to reproduce the analysis.
| Exhibit | File | Code |
|---|---|---|
| Table 1 | NA | NA |
| Table 2 | data/derived/top-industries.csv |
analysis/toplist.do |
| Table 3 | analysis/cost_by_naics2.csv |
analysis/counterfactual.do |
| Figure 1 | NA | NA |
| Figure 2 | NA | NA |
| Figure 3 | text/fig3.pdf |
analysis/scatter_customers_teamwork_byoccupation.do |
| Figure 4 | text/fig4.pdf |
analysis/risk-by-density.do |