Data and Code for: Advanced Technology Adoption: Selection or Causal Effects?

Principal Investigator(s): View help for Principal Investigator(s) Daron Acemoglu, Massachusetts Institute of Technology; Gary Anderson, National Science Foundation. Directorate for Social, Behavioral and Economic Sciences; David Beede, United States Department of Commerce. Bureau of the Census; Catherine Buffington, United States Department of Commerce. Bureau of the Census; Emin Dinlersoz, United States Department of Commerce. Bureau of the Census; Lucia Foster, United States Department of Commerce. Bureau of the Census; Nathan Goldschlag, United States Department of Commerce. Bureau of the Census; John Haltiwanger, University of Maryland; Zachary Kroff, United States Department of Commerce. Bureau of the Census; Pascual Restrepo, Boston University; Nikolas Zolas, United States Department of Commerce. Bureau of the Census

Version: View help for Version V1

Name	File Type	Size	Last Modified
Code			04/03/2023 02:38:PM
Output			04/03/2023 02:38:PM
readme.docx	application/vnd.openxmlformats-officedocument.wordprocessingml.document	23.2 KB	04/03/2023 10:37:AM

Project Citation:

Acemoglu, Daron, Anderson, Gary, Beede, David, Buffington, Catherine, Dinlersoz, Emin, Foster, Lucia, … Zolas, Nikolas. Data and Code for: Advanced Technology Adoption: Selection or Causal Effects? Nashville, TN: American Economic Association [publisher], 2023. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 2023-05-22. https://doi.org/10.3886/E187861V1

Project Description

Summary: This paper studies the employment dynamics for firms that adopt advanced technologies versus firms that do not. The findings suggest that firms that adopt advanced technologies are different from those that do not in terms of their size and growth patterns. Prior to the time period during which the adoption of advanced technologies intensified, these firms were larger and grew faster. However, the adoption of advanced technologies did not lead to significant changes in their employment and growth patterns. These findings suggest that adoption is driven more by selection and have implications on how we assess the impacts of advanced technology adoption on firm performance.

Scope of Project

Subject Terms: Technology Adoption; Firm Dynamics; Robotics; Automation; Artificial Intelligence

JEL Classification:
      L11 Production, Pricing, and Market Structure; Size Distribution of Firms
      L22 Firm Organization and Market Structure
      L25 Firm Performance: Size, Diversification, and Scope
      O14 Industrialization; Manufacturing and Service Industries; Choice of Technology
      O33 Technological Change: Choices and Consequences; Diffusion Processes

Geographic Coverage: National

Time Period(s): 1976 – 2019

Universe: The underlying data consists of a nationally-representative sample of U.S. firms from the Annual Business Survey (ABS), sampled in 2019. The sample is combined with the Longitudinal Business Database (LBD) to assess the firm histories and employment growth patterns.

Data Type(s): census/enumeration data; survey data

Methodology

Response Rate: The 2019 ABS data was collected from June through December 2019. The response rate for the portion of the survey used in this paper was 68.7%

Sampling: The Annual Business Survey (ABS) sampled over 300,000 employer businesses for the reference period of 2016--2018. The set of firms sampled in the technology module was dictated by the general sampling scheme for the 2019 ABS, a primary goal of which is to provide tabulations of collected data by various ownership characteristics. For details on the sampling methodology, see https://www.census.gov/programs-surveys/abs/technical-documentation/methodology.2019.html.

The ABS sampling universe was created using Census Bureau's Business Register administrative data from 2018, which provides the information on industry classification, receipts, payroll and employment for the construction of ABS universe. The ABS universe was stratified by state, frame, and industry, where the frame refers to categories of ownership characteristics for businesses. The Census Bureau used several sources of information to estimate the probability that a business is minority or women-owned. These probabilities were then used to place each firm in the ABS universe to one of nine frames that span key race and ethnicity categories, plus gender and public ownership status. Large companies were selected with certainty based on volume of sales, payroll, or number of paid employees.

Data Source: U.S. Census Bureau's Annual Business Survey (ABS)
U.S. Census Bureau's Longitudinal Business Database (LBD)

Weights: We construct firm weights based on the 2018 LBD to make the sample more representative and account for the response rate. Specifically, we first stratify firms in the 2018 LBD and 2019 ABS by the same size, age, and industry categories (12 size categories, 12 age groups and 19 two-digit NAICS sectors). Each firm in a stratum in the ABS is then assigned the same weight calculated by dividing the firm count in the corresponding 2018 LBD stratum by the firm count in the 2019 ABS stratum. Using the LBD-generated weights gets us much closer to the count and employment distributions found in the 2018 LBD, making our results applicable to the population of employer businesses.

Unit(s) of Observation: Firm-Year

Geographic Unit: National

Related Publications

Download this project

Published Versions

V1 [2023-05-22]

Export Metadata

Dublin Core

DDI 2.5

Report a Problem

Found a serious problem with the data, such as disclosure risk or copyrighted content? Let us know.

This material is distributed exactly as it arrived from the data depositor. ICPSR has not checked or processed this material. Users should consult the investigator(s) if further information is desired.