Table 10.

Robotisation and risk of Covid-19 contagion in the workplace, by remote work.

	Baseline	Add controls	Baseline	Add controls
	OLS	OLS	2SLS	2SLS
	(1)	(2)	(3)	(4)
Robots per 1000 workers, ln $\times$	$-0.{300}^{★★}$	$-0.{388}^{★★}$	$-0.{310}^{★★★}$	$-0.{414}^{★★}$
Remote work = 0	(0.107)	(0.167)	(0.100)	(0.163)
Robots per 1000 workers, ln $\times$	$-0.{053}^{★★★}$	$-0.{090}^{★★★}$	$-0.{050}^{★★★}$	$-0.{070}^{★}$
Remote work = 1	(0.003)	(0.028)	(0.005)	(0.042)
Remote work dummy	$✓$	$✓$	$✓$	$✓$
ICT adoption controls		$✓$		$✓$
Capital intensity		$✓$		$✓$
Observations	259	257	259	257
R-squared	0.216	0.312	0.215	0.311

Notes: The dependent variable is the industry-level risk of Covid-19 contagion in the workplace, as estimated by INAIL (2020). The remote work dummy is based on the remote work index in Barbieri et al. (2020), and takes value 1 if the sector exhibits a level above the country average. The ICT adoption controls include: percentage of firms buying cloud computing services in 2018; percentage of firms in 2019 that use Enterprise Resource Planning software package to share information on sales/purchases with other internal functional areas; percentage of firms in 2019 that use Customer Relationship Management software to collect, file and share data; percentage of firms in 2019 that use Customer Relationship Management software for marketing analysis; percentage of firms that have purchased between 2015 and 2017 goods and services in the area Internet of Things; percentage of workers in 2019 that were provided portable devices allowing internet connection for business purposes. Capital intensity is measured using the logged capital-labour ratio. The 2SLS specifications instrument the log of the number of robots per 1000 workers in Italy in 2018 using the same variable for Japan and South Korea. Standard errors clustered at the aggregate industry level used in the IFR dataset are shown in parentheses. ${}^{★★★}$ indicates coefficients significantly different from zero at the 1% level.