The Covid-19 pandemic has led to the closure of cafés and restaurants, a big increase in home working and some people engaged in panic buying. This has fed through to an increase in spending in supermarkets and food retailers – and driven up grocery prices.
The pandemic has brought higher consumer spending in the grocery sector, as well as presenting challenges to supply chains, and resulting in the need for retailers to implement social distancing in stores. Changes in market conditions, whether it is higher demand or increased costs, typically feed into higher prices. This has been the case with grocery prices over the pandemic. At the beginning of lockdown, on average, prices rose by around 2.5%. This was fuelled by a fall in promotions. These higher prices have persisted since lockdown began.
How can real-time scanner data help us to track the impact of coronavirus on prices?
A burgeoning body of research is using real-time data to understand how Covid-19 is affecting economic activity (for example, see Baker et al, 2020; Bartik et al, 2020; Chetty et al, 2020; Villas-Boas et al, 2020). Unlike many national statistics, real-time data are available with almost no lag and therefore can immediately be informative about the impact of the pandemic and the policy responses.
Much of this work is focused on documenting changes in employment and consumer expenditure. Partly due to data limitations, there is much less work documenting the impact of the crisis on prices (Cavallo, 2020, who reweights the consumer price index measure of economy-wide inflation with Covid-19 spending patterns, is one exception). At the same time, national statistical agencies face challenges in measuring inflation during lockdowns (see Diewert and Fox, 2020).
Related question: How can we measure consumer price inflation in lockdown?
Scanner data, in which a large sample of households record purchases of grocery products, provide a way of tracking what is happening to prices in an important sector of the economy.
This data source has a number of strengths for measuring inflation in the grocery sector. It contains information on millions of transaction prices and consumer expenditures for detailed products (for example, Ivancic et al, 2011); and it allows researchers to track household-specific inflation rates (for example, Kaplan and Schulhofer-Wohl, 2017).
Scanner data also have the advantage of being available in real-time. This means that they can be used to provide a detailed picture of what is happening to grocery prices over the crisis.
What has happened to grocery prices?
Figure 1 uses scanner data for the UK from Kantar FMCG Purchase Panel (‘fast-moving consumer goods’). It shows how the average price of grocery products has changed from the beginning of the year until 17 May, and how this compares with preceding years.
The average price in 2020 was stable until the week when lockdown began, at which point, it jumped by over 2.5%. The increase has persisted in the weeks that followed. Figure 1 shows that over the same period in 2016, 2018 and 2019, prices were stable. In 2017, prices rose steadily from the third week of the year, but the overall increase in the average price was less than half of the rise at the start of lockdown.
Figure 1 shows that the average price rose at the point of lockdown. Jaravel and O’Connell (2020) show that if product level price changes are weighted by their importance in households’ grocery baskets, a similar picture emerges. In other words, these higher prices translated into higher grocery inflation.
Figure 1: Average price for grocery products
Source: Institute for Fiscal Studies analysis using Kantar Worldpanel
What has happened to promotions?
Typically, around 28% of grocery transactions involve some sort of promotion – for example, ‘buy one, get one free’, 33% free or a discounted price. Figure 2 shows how the share of grocery transactions involving promotions has changed in 2020, as well as how it changed over the same period in 2019.
In the first 11 weeks of both 2019 and 2020, the share of transactions involving promotions was stable at 28%. But by the beginning of lockdown, the share had dropped to less than 24%, and this lower level has persisted. In contrast, in 2019, the share of promotion transactions remained around 28% over the whole period.
This reduction in promotions has been key in fuelling higher grocery prices. Jaravel and O’Connell (2020) show that grocery price inflation at the beginning of lockdown, when computed only using non-promoted transactions, is about 1% less than half when promotions are taken into account.
The reduction in transactions involving promotions is a key difference with what happened in the 2008/09 downturn, during which consumers purchased more groceries on sale (Griffith et al, 2016).
Figure 2: Share of grocery transactions on promotion
Source: Institute for Fiscal Studies analysis using Kantar Worldpanel
Do these price rises indicate profiteering?
The UK prime minister expressed concern about profiteering during his daily coronavirus briefing on 25 March, with many stories in the newspaper about high prices. Higher prices in the grocery market could reflect increases in retailers’ profit margins. But they may also reflect increased costs, associated with disrupted supply chains and production, and the need for businesses to implement social distancing among their employees.
Related question: What should we do about price gouging?
What is the longer-run context for food prices?
Figure 3 puts the recent grocery price rises into a longer-run context by showing how food and non-alcoholic drinks prices have evolved over the 30-year period from 1988 to 2019. Based on the Office for National Statistics (ONS) consumer price index, this shows how the food and non-alcoholic drink price index has changed relative to the price index for all consumer goods and services.
Between 1988 and 2016, food became less expensive relative to all goods and services, with the relative price of food more than 10% lower by 2006. In contrast, from 2006 to 2013, relative food prices rose by over 14%. This was then followed by another period of falling relative food prices. Over the whole period 1988-2019, relative food prices fell by 5.5%.
Therefore, while the increase in grocery prices at the beginning of lockdown was significant (it resulted in more grocery inflation in a month than we typically see in year), it does come after six years of falling relative food prices.
It remains to be seen how grocery prices will evolve as we emerge from lockdown: in particular, will the price spike driven by fewer promotions be reversed, will it persist or is it a sign of the return of food price inflation.
This will depend in part on whether grocery demand returns to normal, and whether there is further disruption to supply chains. It will also depend on what happens to the value of sterling and to trade barriers as the UK emerges from the Brexit transition period.
Related question: Should Brexit trade policy changes be delayed because of Covid-19?
Figure 3: Relative price of food and non-alcoholic drinks
Source: Office for National Statistics, Consumer Price Index
Where can I find out more?
Xavier Jaravel and Martin O'Connell show that grocery prices rose by 2.4% in a month at the beginning of lockdown, fuelled by a fall in promotions.
Richard Blundell, Rachel Griffith, Peter Level and Martin O’Connell discuss challenges facing the Office for National Statistics in measuring inflation during lockdown.
Dirk Broeders, Gavin Goy, Annelie Petersen and Nader de Vette discuss how the value of financial instruments can be used to infer market-based inflation expectations following the Covid-19 shock.
Oliver Blanchard talks about the likelihood of there being deflation or inflation following the pandemic.