The Centre has extended the lockdown period till 30 June, with updated guidelines put in place to ease the gradual transition of opening up the economy. Even as we move into Lockdown 5.0 (or Unlock 1.0, as many are calling it), there is still little evidence on how the social mobility and distancing has changed across these many different phases of lockdowns. Earlier, on 12 May 2020, Prime Minister Narendra Modi had said that "lockdown 4.0 will be different from the previous ones". The comment had attracted a lot of curiosity. The underlying rationale for the government's strategy was that since coronavirus is expected to stay for a while, we must start adjusting our lives around it and kickstart economic activity by relaxing certain restrictions.
Many, however, have argued that the restrictions were not "very different" from the previous lockdowns. There are already many debates on the effect of the lockdown on the COVID-19 outbreak in India; we do not wish to get into those debates in this article. We attempt to answer a more fundamental question: How effective was the lockdown in terms of compliance across different phases on the ground? In other words, how has public mobility been impacted as a result of these shutdowns over time? To what extent did the crowding reduce as a result of it?
Measuring the level of compliance of a nationwide lockdown is inherently difficult. There was detailed media coverage of the ground reality following the announcement of the lockdown: the reports ranging from places showing high to low compliance. It is fairly obvious that making a general claim about compliance based on media reports is problematic, as it picks particular incidents and biases the sample in multiple ways. We need systematic data to make a general claim about its effectiveness.
Thus, we used anonymised location data aggregated at the country-level from phones released by Google as an indicator for the level of social distancing during the different phases of lockdown. For more details on how the data is collected, you can visit the Google mobility website. One obvious advantage of this data is that it is systematic in nature. Second, it also provides a baseline, which enables us to compare pre-lockdown and post-lockdown situations. The baseline is the median value for the corresponding day of the week during the 5-week period (3 January"6 February). In our view, this provides a better basis to measure and compare the level of social distancing.
The first big lockdown announcement was made by Modi when he exhorted people to observe a 'Janata Curfew' on Sunday, 22 March. According to the public mobility data, there was a very high degree of compliance seen on this day. Figures 1-5 show a sharp drop in the number of people visiting retail and recreation spaces, groceries and pharmacies, parks, transit stations and workplace. It also shows an increase in people staying at homes. Note that residential changes are less compared to other indicators because residential change is measured by change in duration, while the other indicators are measured as change in number of visitors.
After 22 March, as would be expected, we see a sharp increase in mobility with more people visiting retail and recreation spaces, parks, groceries and transit stations. However, interestingly, the mobility to work still went down and the amount of time spent at homes increased. This suggests that even before the complete lockdown was announced, people had started avoiding workplaces and were staying at home.
The phase-I lockdown was put in place from 25 March with an almost complete shutdown of all services and factories. All indicators show a sharp fall in the number of visitors and an increase in the duration of people staying at home. Post that, the lockdown 2.0 and 3.0 announcements have made several relaxations, but these have not led to any significant changes in mobility across various indicators. Rather, after the announcement of lockdown 2.0, there was a sudden drop in visitors at retail places and parks.
Movement to grocery stores and pharmacies, regardless of the announcements, has been going up steadily. This could be the result of falling supplies of daily essentials and medicines at home, which are forcing people to come out of their houses despite severe movement restrictions.
Modi had said that the fourth phase of the lockdown would be different from the ones before it. The prime minister had said that we need to find a balance between "fighting coronavirus and moving forward". However, if data is anything to go by, the changes in guidelines have not led to a lot of change in mobility levels.
A strict lockdown without opening up any economic activity may not be feasible anymore. The recently-released quarter 4 GDP numbers show that the Indian economic growth is at 4.2 percent, even though it accounts for only one week of the lockdown phase (25 March to 31 March). This already is an 11-year low for India. On the back of an already precarious economic position, the coronavirus-induced shutdowns will only lead to deeper economic troubles for India. Some estimates suggest that for every day that the lockdown persists, the economy suffers a loss of Rs 35,000-Rs 40,000 crores.
In the latest announcement by the MHA, the lockdown has been extended to 30 June. All inter-state and intra-state travel restrictions are to be lifted by 1 June; and malls, restaurants, and religious places will be allowed to open from 8 June in some states. These measures are likely to encourage an increase in mobility across the country. However, only time will tell how much the announcement can really allay the uncertainty and concerns about possible risks, as a result of which activity revival may still take time.
Although mobility data provides insights, it comes with certain caveats that must be pointed out. Since the data is collected from the mobiles phones with GPS, it leaves out a significant section of the population " people without GPS-enabled phones or people with no phones. Hence, some incidents could systematically be missed by this data: such as situations where migrant workers were forced to congregate for food distribution, registration for travel, etc. Therefore, the data over-represents the mobility patterns of the reasonably well-to-do segment of the population. Notwithstanding the limitations, we believe it is still a better indicator than other publicly available alternatives, such as media reports.
Fahad Hasin studies political science at Ashoka University. Ishaan Bansal studies economics at Ashoka University.