One Question Quiz
The streets of lockdown (photo: The Spinoff)
The streets of lockdown (photo: The Spinoff)

ScienceOctober 24, 2020

When it all stopped: measuring the impacts of the great lockdown experiment

The streets of lockdown (photo: The Spinoff)
The streets of lockdown (photo: The Spinoff)

In 2020, the Covid-19 pandemic prompted New Zealand and much of the world to undertake something few of us had ever contemplated: a near-total lockdown of society. In this Lockdown legacies series, James Dann explores the impacts of those extraordinary measures, intended and otherwise. Today, part one: the lockdown halt.

This project was made possible thanks to support from the Aotearoa New Zealand Science Journalism Fund.
 
In the early stages of the pandemic going global, as Covid-19 tore across Europe and North America, one meme spoke to the moment. Nature is healing – we are the virus. It was originally, sincerely deployed as wildlife returned to places it had long since departed – sea life in Venice’s canals, goats taking over Welsh villages. While much of it, like most good things on the internet, turned out not to be true, there was a kernel of truth in the initial premise. With the world shutting down and human interactions contracting in a way not seen in modern history, the human impact on the world was severely reduced. While we might not have been able to observe dolphins in the waters of Venice, there were an endless number of other observations that could be made, from the completely trivial through to the deeply meaningful. In a way, lockdown was a huge experiment – if you cut human activity, what are the results?

In this series, I’ll be looking at some of the fascinating research that came out of the lockdown. Not the science of the coronavirus itself, but the indirect effects on all sorts of facets of society that emerged from the great retrenchment of human activity. The absence of human life is a great way to measure how we live. In part two, I’ll examine the impacts on the world around us, including air quality, noise pollution, and measures of water quality. Next, I’ll survey some of the non-Covid health effects of the lockdown, from reductions in the number of people getting other diseases, to fewer workplace accidents and traffic deaths. Finally, I’ll take a dive into the research on the societal changes of the lockdown, including impacts on sleep, exercise, relationships, and mental health. 

How locked were we?

But before we draw any conclusions about the great stoppage, we need to first establish just how locked down we were. To what extent did our activity drop during the alert level four lockdown? Was the shutdown felt evenly across different regions, and different areas of human activity?

As reported by the Spinoff back in April, Google used location data to analyse changes all around the world, including in New Zealand. These data showed some large changes in movements, including a 91% decrease in retail and recreation activity, use of parks down by 78%, and visits to grocery stores and pharmacies down by 54% in the first week of the level four lockdown across the country. The number of people at their places of work was down 59%, and transit station usage was down 84%. With all these reductions, people had to be going somewhere – and this was seen with a 22% increase in residential activity.

The data provided by Google was drawn from people who had turned location services on in Google Maps on their phone. As Google is close to omnipresent around the world, this information is most useful for comparing New Zealand’s lockdown to other nations’ lockdowns. However, as it requires people to opt-in to allow Google to use their data, it only provides information from a small subset of people, which may not be representative of the wider population. In some areas, especially the regions, there aren’t enough data points to provide a robust analysis.

A larger and more inclusive dataset was created by DataVentures, the commercial wing of StatsNZ. Late last year, they had announced that they would be collecting anonymous data from cellphones to help make decisions around infrastructure projects, or plan for emergencies (for those worried about potential privacy implications, I asked the Office of the Privacy Commissioner whether they had any issues with this collection of data, and they said that they could “express comfort that the methodology and processes that have been put in place by Data Ventures are robust and privacy enhancing”.) 

Only six months later that data proved itself very useful for analysing movement patterns in an unprecedented national emergency. The two main telcos, Vodafone and Spark, provide Stats NZ with the number of mobile devices in a suburb each hour of the day, aggregated and stripped of any metadata that could be used to identify the user.

When we stopped moving

DataVentures brought the information together in a National Mobility Index, which creates “population estimates of residents and visitors in New Zealand every hour down to suburb level”. The index supports the findings of both the Google mobility data, and the observations made by anyone with eyes – that there was dramatically reduced mobility around the country during lockdown.

National Mobility Index

This reduction was about 50% across the country during the six weeks of level four. That is, people all over the country were making about half as many movements as they would have done under normal conditions. As the levels dropped, mobility increased. In level three, there was more mobility than level four, and more again in level two, as we returned to something close to normal.

The data was also able to be split in a number of ways. There was a massive reduction in the amount of activity at retail stores and workplaces. And there were regional differences. The biggest drops in mobility were seen in the urban centres, while in more rural areas, such as Northland and the West Coast, there was still a drop, but it was not quite as pronounced. One explanation for this difference may be that a higher proportion of people in rural areas were working on or linked to farms, which were essential services, and thus their mobility didn’t change much. 

The team from DataVentures also observed that those in rural areas were likely to need to travel “to denser populated areas where points of interest such as supermarkets reside, hence the smaller decrease in mobility”, and, to a lesser extent, “the fact devices in rural areas are harder to pick up so the change in counts for an area may experience less change as a result”.

The mobility data also showed a big drop in transit activity. This was supported with a huge drop in the amount of fuel being used in the country, as measured by the Z Energy fuel supply index. Prior to lockdown, Z was supplying around 40 million litres of fuel a week; by April 5, the supply was about a quarter of that. 

Another series that shows the effects of the lockdown is New Zealand’s best named statistic, the ANZ Truckometer. For both heavy and light trucks, the April reading shows a huge plunge. It is worth noting how quick the Truckometer bounced back, with new highs in both June and July. These drops not only shows that we were largely staying put, but the reduction in fuel supply means a reduction in fuel usage, which leads to changes in air pollution and carbon emissions that we’ll discuss in subsequent pieces in this series.

While this might all seem bleedingly obvious, the ability to quantify these observations is critical for turning anecdata into real evidence. New Zealand’s strict level four lockdown led to a massive drop in our usual activities outside of the home – going to school, work, and the shops. It reduced economic activity, fuel consumption, and movement around the country. And with this unprecedented contraction of activity, there were many flow-on effects that have resulted in some interesting findings that we will cover in the next three parts of this series, which continues tomorrow.

Keep going!