For whom the bell tolls: The firm-level effects of automation on wage and gender inequality

G. Domini; M. Grazzi; D. Moschella; T. Treibich

doi:10.1016/j.respol.2022.104533

For whom the bell tolls: The firm-level effects of automation on wage and gender inequality

G. Domini, M. Grazzi, D. Moschella, T. Treibich^*

^*Corresponding author for this work

Research output: Contribution to journal › Article › Academic › peer-review

Abstract

This paper investigates the impact of investment in automation-and AI-related goods on within-firm wage inequality in the French economy during the 2002-2017 period. We document that most wage inequality in France is accounted for by differences among workers belonging to the same firm rather than by differences between sectors, firms, and occupations. Using an event-study approach on a sample of firms importing automation-and AI-related goods, we find that spike events related to the adoption of automation-and AI-related capital goods are not followed by an increase in within-firm wage inequality or in gender wage inequality. Instead, wages increase by 1% three years after the events at different percentiles of the distribution. Our findings are not linked to the rent-sharing behavior of firms obtaining productivity gains from automation and AI adoption. Instead, if wage gains do not differ across workers along the wage distribution, worker heterogeneity will still be present. Indeed, in agreement with the framework in Abowd et al. (1999b), most of the overall wage increase is due to the hiring of new employees. This adds to previous findings presenting a picture of a 'labor friendly' effect of the latest wave of new technologies within adopting firms.

Original language	English
Article number	104533
Number of pages	30
Journal	Research Policy
Volume	51
Issue number	7
DOIs	https://doi.org/10.1016/j.respol.2022.104533
Publication status	Published - 1 Sept 2022

JEL classifications

j16 - "Economics of Gender; Non-labor Discrimination"
j31 - "Wage Level and Structure; Wage Differentials"
l25 - Firm Performance: Size, Diversification, and Scope
o33 - "Technological Change: Choices and Consequences; Diffusion Processes"

Keywords

Automation
AI
Wage inequality
Gender pay gap
TECHNOLOGICAL-CHANGE
COUNTRIES EVIDENCE
EMPIRICAL-EVIDENCE
POLARIZATION
EMPLOYMENT
INVESTMENT
WORKERS
GROWTH
ROBOTS
TRADE

Access to Document

10.1016/j.respol.2022.104533Licence: CC BY

Cite this

@article{5833eab8939c441fa48252b2cbb636cb,

title = "For whom the bell tolls: The firm-level effects of automation on wage and gender inequality",

abstract = "This paper investigates the impact of investment in automation-and AI-related goods on within-firm wage inequality in the French economy during the 2002-2017 period. We document that most wage inequality in France is accounted for by differences among workers belonging to the same firm rather than by differences between sectors, firms, and occupations. Using an event-study approach on a sample of firms importing automation-and AI-related goods, we find that spike events related to the adoption of automation-and AI-related capital goods are not followed by an increase in within-firm wage inequality or in gender wage inequality. Instead, wages increase by 1% three years after the events at different percentiles of the distribution. Our findings are not linked to the rent-sharing behavior of firms obtaining productivity gains from automation and AI adoption. Instead, if wage gains do not differ across workers along the wage distribution, worker heterogeneity will still be present. Indeed, in agreement with the framework in Abowd et al. (1999b), most of the overall wage increase is due to the hiring of new employees. This adds to previous findings presenting a picture of a 'labor friendly' effect of the latest wave of new technologies within adopting firms.",

keywords = "Automation, AI, Wage inequality, Gender pay gap, TECHNOLOGICAL-CHANGE, COUNTRIES EVIDENCE, EMPIRICAL-EVIDENCE, POLARIZATION, EMPLOYMENT, INVESTMENT, WORKERS, GROWTH, ROBOTS, TRADE",

author = "G. Domini and M. Grazzi and D. Moschella and T. Treibich",

note = "data source: D{\'e}claration Annuelle des Donn{\'e}s Sociales (DADS), a confidential database provided by the French national statistical office (INSEE) and based on the mandatory forms that all establishments with employees must submit to the social security authorities. The second source is the transaction-level international trade dataset by the French customs office (Direction G{\'e}n{\'e}rale des Douanes et des Droits Indirects, DGDDI), containing detailed information on import and export flows. We also use FICUS and FARE, two private datasets provided by INSEE, which are based on the fiscal statements that all French firms must make to the tax authorities and which contain detailed balance sheet and revenue account data. ",

year = "2022",

month = sep,

day = "1",

doi = "10.1016/j.respol.2022.104533",

language = "English",

volume = "51",

journal = "Research Policy",

issn = "0048-7333",

publisher = "Elsevier",

number = "7",

}

TY - JOUR

T1 - For whom the bell tolls: The firm-level effects of automation on wage and gender inequality

AU - Domini, G.

AU - Grazzi, M.

AU - Moschella, D.

AU - Treibich, T.

N1 - data source: Déclaration Annuelle des Donnés Sociales (DADS), a confidential database provided by the French national statistical office (INSEE) and based on the mandatory forms that all establishments with employees must submit to the social security authorities. The second source is the transaction-level international trade dataset by the French customs office (Direction Générale des Douanes et des Droits Indirects, DGDDI), containing detailed information on import and export flows. We also use FICUS and FARE, two private datasets provided by INSEE, which are based on the fiscal statements that all French firms must make to the tax authorities and which contain detailed balance sheet and revenue account data.

PY - 2022/9/1

Y1 - 2022/9/1

N2 - This paper investigates the impact of investment in automation-and AI-related goods on within-firm wage inequality in the French economy during the 2002-2017 period. We document that most wage inequality in France is accounted for by differences among workers belonging to the same firm rather than by differences between sectors, firms, and occupations. Using an event-study approach on a sample of firms importing automation-and AI-related goods, we find that spike events related to the adoption of automation-and AI-related capital goods are not followed by an increase in within-firm wage inequality or in gender wage inequality. Instead, wages increase by 1% three years after the events at different percentiles of the distribution. Our findings are not linked to the rent-sharing behavior of firms obtaining productivity gains from automation and AI adoption. Instead, if wage gains do not differ across workers along the wage distribution, worker heterogeneity will still be present. Indeed, in agreement with the framework in Abowd et al. (1999b), most of the overall wage increase is due to the hiring of new employees. This adds to previous findings presenting a picture of a 'labor friendly' effect of the latest wave of new technologies within adopting firms.

AB - This paper investigates the impact of investment in automation-and AI-related goods on within-firm wage inequality in the French economy during the 2002-2017 period. We document that most wage inequality in France is accounted for by differences among workers belonging to the same firm rather than by differences between sectors, firms, and occupations. Using an event-study approach on a sample of firms importing automation-and AI-related goods, we find that spike events related to the adoption of automation-and AI-related capital goods are not followed by an increase in within-firm wage inequality or in gender wage inequality. Instead, wages increase by 1% three years after the events at different percentiles of the distribution. Our findings are not linked to the rent-sharing behavior of firms obtaining productivity gains from automation and AI adoption. Instead, if wage gains do not differ across workers along the wage distribution, worker heterogeneity will still be present. Indeed, in agreement with the framework in Abowd et al. (1999b), most of the overall wage increase is due to the hiring of new employees. This adds to previous findings presenting a picture of a 'labor friendly' effect of the latest wave of new technologies within adopting firms.

KW - Automation

KW - AI

KW - Wage inequality

KW - Gender pay gap

KW - TECHNOLOGICAL-CHANGE

KW - COUNTRIES EVIDENCE

KW - EMPIRICAL-EVIDENCE

KW - POLARIZATION

KW - EMPLOYMENT

KW - INVESTMENT

KW - WORKERS

KW - GROWTH

KW - ROBOTS

KW - TRADE

U2 - 10.1016/j.respol.2022.104533

DO - 10.1016/j.respol.2022.104533

M3 - Article

SN - 0048-7333

VL - 51

JO - Research Policy

JF - Research Policy

IS - 7

M1 - 104533

ER -