The persistent effects of COVID-19 on labour outcomes: evidence from Peru

ABSTRACT

This paper takes a new look at the impact of COVID-19 on labour outcomes more than a year after the beginning of the pandemic. We use a labour survey from Lima, Peru, with monthly data from January 2019 up to June 2021. We corroborate the early dramatic impact of the pandemic documented in other countries. These initial effects attenuate over time, but persist and remain sizable: even by mid-2021, there is a reduction of almost 20% in hours worked and labour income. Our findings highlight the limitation of policy changes and societal adaptations to ameliorate the economic impact of COVID-19.

KEYWORDS:

• COVID-19
• employment
• labour income
• developing countries

JEL CLASSIFICATION:

• J21
• O12

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

¹ The mortality rate since the beginning of the pandemic is around 6,000 deaths per million inhabitants, a rate twice as large as the next highest one and above countries like Brazil and India. See Figure C.1. in the online appendix.

² See online appendix A for further details.

³ Sampling is done at dwelling level. Every month 1,600 dwellings are programmed to be surveyed. Before the pandemic, the survey response rate was around 87%. This rate of data collection continued almost normally during the pandemic, except for March 2020 when the response rate dropped to 55%, but recovered the next month. We observe similar pattern in the number of observations. See Figure C.4 in the online appendix.

⁴ A block is the primary sampling unit, and the only geographic identifier provided in the dataset. There are 51 municipalities in the Lima Metropolitan area, while our sample covers around 3,900 blocks.

⁵ See Figure C.3 in the online appendix for the monthly evolution of our main outcome variables from January 2019 to June 2021.

⁶ This might have occurred because they were more likely to work in affected industries or unable to work remotely. See, for instance, Baek et al. (2021); Bartik et al. (2020); Betcherman et al. (2020); Cho and Winters (2020); Couch et al. (2020); Genoni et al. (2020), Hoehn-Velasco, Silverio-Murillo, and de la Miyar (2021); Jain et al. (2020); Kikuchi, Kitao, and Mikoshiba (2021).

⁷ We cannot use the last quarter because individuals in the panel sample are interviewed in yearly intervals. Thus, there would not be pre-pandemic data for individuals interviewed after March 2021.

⁸ For instance, the reduction of labour income for an average informal worker without high school was 76%. In contrast, the reduction for a formal, more educated worker was only 40.8% ($-0.759+0.159+0.192$).

⁹ Note that most schools in Peru have remained closed for in-person classes since the beginning of the pandemic through August 2021.

¹⁰ See Table D.5 in the online appendix.