Off the chart: Australians were world leaders in panic buying, beating UK and Italy
Australians lead the world in panic buying in the early months of the coronavirus pandemic, according to new research that has analysed Google searches, including those for toilet paper and supermarket opening hours.
Researchers from the University of New South Wales said they had to build a new scale for graphs showing Australian panic, as levels were four to five times higher than countries like the UK and Italy, who were worse-hit by Covid-19.
Dr Tim Neal said that unlike in other countries, the spikes of Australian panic buying were not aligned to either government announcements or rising Covid-19 cases, meaning that the Australian media’s coverage of hoarding could have contributed to the world-leading levels of panic.
Neal and Prof Mike Keane from the UNSW business school developed an index to measure panic buying across 54 countries during the pandemic. The researchers analysed Google search data for words like “toilet paper”, “panic buying”, “hoarding”, “supermarket”, “recession” and “unemployment” in various languages.
Related: Rolling in the aisles: Australians turn to TikTok to mock toilet paper panic-buying
Their report found that Australia was “notable for the incredible speed and scale with which panic took hold in early March,” backed up by sales data that shows supermarket spending rose 24.1% that month.
Australia peaked at a panic level of 0.79 on the research’s index, while the UK peaked at approximately 0.175, Italy at 0.15, and France at 0.09.
“When you look at March and April, Australia was far and away the most panicked country,” Neal said. “We haven’t calculated who came second, who came third – it wasn’t a close contest. Japan had a big spike but it very quickly disappeared.
“At the peak of the crisis at Australia, 0.8% of all Google searches included one of our key words related to panic. That doesn’t sound like a lot, but when you consider that people can search anything, 0.8% of people searched for this one specific thing is quite remarkable. Prior to March, in February and January, these terms would only get 0.005%.”
The research also found that the spike in panic could not be explained just by infection rates and government lockdowns.
“Unlike in other countries, the escalation in panic does not appear to correspond with any significant increase in domestic Covid-19 cases,” it said. “Nor does any important policy announcement seem to explain the panic: While a travel ban from Iran was announced on 1 March, it seems unlikely that this was important enough to be the direct cause.”
Neal told Guardian Australia that this meant the Australian media could have played a part.
“The way that videos of people fighting over toilet paper took off in the Australian media could be a factor. What we do in our paper is we not only measure panic, but we try and model it and try and understand why it increased when it did.
“We broadly considered two things that could cause panic: the transmission of the virus and government announcements. We have a good way of explaining why most counties behaved the way that they did. But Australia was one of the few where the model did a really bad job.”
A spokeswoman for Kleenex Australia said that supermarket shelves were “back to normal” after the panic of earlier months.
“It has certainly been a busy few months,” she said. “While supermarket shelves are back to normal, we are still operating at full capacity to ensure healthy stock levels across the country.”
Neal said the researchers looked at Google data as a way to complement traditional sales data to measure panic. Many countries do not produce sales data, and online search data allowed them to compare across 54 different countries, on a day-to-day scale.
“The use of Google allowed an international comparison,” he said. “And also showed us the speed at which panic buying set in.”
“The traditional way is to collect sales data from supermarkets. The advantage of that is you can look at specific products and get a clear sense how much demand surged for specific goods. But it is not available for many counties. The other disadvantage is it is usually monthly data.”
