We’ve been doing a lot of internal research here at the Social Economy Data Lab to complement the work of our partners at Tortoise Media. Here on the SEDL site we’ll be drilling down into what the data can tell us about the social economy in particular, and areas that will need more focused and concentrated attention from grant-funders and social investors.
‘Left behind’ communities had already been identified as economically vulnerable prior to the corona crisis. They have low levels of social mobility, lows skills or declining industries, and have been excluded from external investment by the public and voluntary sectors. We are concerned that the COVID-19 crisis has the potential to further trap these places in spiralling economic decline.
So we turned to the data to compare left behind economies to a sample of more affluent places, and understand if our concerns were well-founded. It is important to us to use the predictive power of data in response to the pandemic– we need to understand leading economic indicators if we are to help to inform the direction of response efforts to mitigate economic fallout before it happens.
Leading economic indicator set 1: Consumption, proxied by card transactions
Compared to more affluent areas, left behind places have benefited from a larger increase of grocery sales compared to spend for the same week last year (22.55%) and although a very large negative sales effect on all ‘other’ (i.e. non-grocery) sales of 51.67%, this was slightly better than the decreases experienced in more affluent areas when compared to their sales to the previous year.
If we look at Figure 2 above, we can see the £ value change in sales (grocery, ‘other’, and the total) due to the effects (direct and indirect) of COVID-19.
We do need to point out that our card transaction dataset is only about 13%-33% of the card universe (depending area) and that cash sales would not be reflected, so although we are interested in the actual value of sales, it is unhelpful to use these as the true value of income generated in these two area groups. What is interesting, and relevant to our analysis however, is the year on year changes. Both groups benefitted from about a £1Million increase in weekly grocery sales compared to what was spent in the same week last year. It is interesting to reflect that this absolute figure is similar across both groups, despite the fact that affluent areas have larger total sales spends. I.e. this shows us what we expected: people in left behind places are spending larger proportions of their incomes on necessities like groceries (see Figure 3 below).
Left behind places, with lower economic activity before the health crisis, also have much less income to circulate within the local merchants in their communities. In the weekly sales figures for 2019, total sales in more affluent areas were 2 times (1.97 to be precise) the sales generated in left behind communities. Given the benefits of economic multiplier effects, we know that having more money in a local economy can help in supporting economic uplift in the recovery period; and this inequality in (likely ability to) spend is concerning from that perspective.
To fully unpack this, we then also consider changing purchasing power within local economies. Our concern is that while more affluent areas have seen falls in spending during the crisis, they would have a higher purchasing power within their communities to essentially ‘jumpstart’ the economy with increased consumption when the crisis lessens.
Leading economic indicator set 2: Individuals seeking information on benefits qualification
As we are interested in leading economic indicators, we have chosen to use data on benefits calculator use. Seeking information on whether or not people qualify for benefits, would be the first step for many in their pathways to accessing them. It also gives us a sense of where we might expect to see dips in future purchasing power due to reduced incomes.
It is important to note again, that about 250,000 individuals accessed the Turn2us benefit survey dataset in the last week of March 2020, this is not nationally representative dataset as many will be accessing benefits information through other sources. The absolute figures are then less important to focus on; the relevance to our analysis is again the change between time periods. For the benefits data, there was an important trigger point on the 17th March when the Chancellor of the Exchequer announced crisis-related benefits changes. Figure 4 below compares users over a 2 week time period – either side of the announcement on the 17th March 2020.
There was a nation-wide increase of benefit calculator users, with more affluent areas seeing a much larger uptick in individuals seeking information – 247% increase compared to a 133% increase in users from left behind places. Importantly however, the numbers of users from left behind places (volumes wise) has always been significantly larger than users from affluent areas, and this continues to be the case following the benefits announcement.
Despite the increase in users, the percentage of those qualifying remained stable – within a band of 42% to 49% across groups and time-periods. Essentially, in numbers, this means that between 12-25 March, 8,128 individuals had results which indicated they qualify for benefits. An increase of 157% compared to those qualifying between 1-14th March.
What do these indicator set tell us?
From the data above, we can see a large and significant effect on both consumption and benefit calculator usage across both affluent and left behind places. It is important to continue to track this data as the crisis unfolds. Affluent areas have seen a larger decrease in sales, and a bigger increase in benefits calculator uses; they have seen bigger fluctuations to our key metrics due to the crisis. While left behind places may not have seen as much volatility in comparison, they were already spending larger proportions on necessity goods like groceries, have 2 times less income in circulation within local economies, and had larger absolute volumes of individuals accessing the benefits calculator.
As we continue to track these metrics as they unfold over the next 6 weeks (both here and with Tortoise), the critical questions for us are:
- will affluent areas manage to ride out the volatility in the short-to-medium term to return to some level of pre-corona economic stability,
- and will continued and sustained economic pressure on the already weakened economies of left behind places mean that they will fall into deeper recessions?
- And most importantly, in the face of a perfect storm brewing, what can the social economy do, now, in response?
There are two underlying data sources which have fed into our ongoing analysis –
Imfoco: Card transaction, or spend, data
Imfoco compiles, cleans and adds to consumer transaction spend raw data from 6 different card and bank sources. They have transactional data on spend by customer through card and online types (including direct debits). We have used the card transaction spend data to understand where consumers spend. A social purpose entity, they support socially relevant projects with their available datasets.
Turn2us: Benefits calculator dataset
A benefits calculator online helps
individuals to establish whether or not they classify for benefits, and which
ones. A national charity, they aim to provide practical help to those
struggling financially. The benefits calculator data is daily, and has been
provided to us at ward level to protect personal information of data subjects. https://benefits-calculator.turn2us.org.uk/AboutYou
 OCSI and Local Trust,’ Left Behind? Understanding communities on the edge’, 2019 (access @ https://localtrust.org.uk/wp-content/uploads/2019/08/local_trust_ocsi_left_behind_research_august_2019.pdf)