The relationship between mobile broadband usage and user mobility with lockdown restrictions in Spain

ABSTRACT

This paper studies how COVID-19 lockdown restrictions encouraged and allowed people to form habits of increased mobile phone usage in Spain, even after the most restrictive measures were lifted. We have used data from the mobile network of a national telecommunications operator to study the influence of 15 different mobility restrictions on citizens’ use of mobile phones from 15 March 2020 to 15 April 2021 in a big Spanish city. In order to approach this issue, a quasi-experimental technique (before and after methodology) was implemented. We found that the use of smartphones presented an initial reduction with the first and most drastic mobility restrictions, but showed a growing use after their removal. However, user mobility was negatively influenced by all the mobility restrictions.

KEYWORDS:

• COVID-19
• mobility
• internet use
• lockdown
• government restrictions

JEL:

• C01
• O30

DATA AVAILABILITY

The dataset used in this study is exclusive due to confidential agreements.

DISCLOSURE STATEMENT

No potential conflict of interest was reported by the authors.

Notes

1 For instance, the Spanish region of Conil de la Frontera has applied this methodology; see https://www.segittur.es/blog/sin-categorizar/casos-de-uso-del-big-data-en-el-sector-turistico-durante-el-covid-19-el-ejemplo-de-conil-de-la-frontera/.

2 The name of the operator is not shown in the text for reasons of confidentiality.

3 In order to better understand secular trends in average downloaded mobile data from Figure 1, the date were decomposed into trend (see Figure A1 in the supplemental data online) and irregular component (see Figure A2 online). To the extent that average user mobility presents a slight secular trend, this was only done for average downloaded mobile data.

4 It could be interesting to analyse whether people failed to shift back from online after the treatments; unfortunately, we do not have a counterfactual for this kind of analysis (e.g., certain postal codes with absence of COVID or restrictions).

5 Downloaded data analysis may be more potentially compelling than mobility analysis to the extent they are less conditioned by COVID-19 transmission than mobility.

6 A square trend was also tested, but it was not significant.

 

Additional information

Funding

This work was partly supported by FEDER [research project PY20-00228], the Ministerio de Ciencia e Innovación [research project PID2020-119471RB-I00], the Andalusian Regional Government [grant number SEJ-645] and the training received from the University of Malaga Programa de Doctorado en Economía y Empresa (PhD Program in Economics and Business).