Key findings using the latest data include:

• A few local authorities are still seeing rapidly rising case numbers. Blackpool (ranked 9^th, up from 13^th) has seen cases rise from 193 per 100,000 people last week to 271 this week. Similarly, Hartlepool (ranked 24^th, up from 41^st) has seen a dramatic rise from 101 cases last week to 177 cases this week. As a result of increasing case numbers and a high level of underlying systemic risk, both local authorities have jumped up the rankings and are amongst the 25 riskiest places.
• There is relative stability in the level of risk facing local authorities with 8 out of the 10 riskiest places retaining their position at the top of the Index. This stability is driven by a broad slowdown in the number of new cases.
• Our analysis suggests that the ability to work from home is an important part of the explanation for different local authority growth rates in case numbers over the last month. Those local areas with a higher proportion of occupations where people can work from home, have seen a slower rise in cases post lockdown.

Commentary on the results

While the “lockdown” came into force just over a month ago, some areas have seen case numbers rise much faster than others during April and we have documented particularly acute rises across parts of the North of England. Given that the public health restrictions on social distancing are nationwide, this has been a puzzling feature of the last few weeks.

While data on case numbers is far from perfect and likely to be subject to the effectiveness of local testing, our analysis suggests that those places with a higher proportion of people able to work from home, have seen lower increases in case numbers. This finding is robust to the inclusion of other factors such as the number of cases at the beginning of April and local authority population size[1]. In Middlesbrough, which has consistently been one of the riskiest local authorities in the Index, our calculations suggest that only around 20% of the workforce are in jobs that can be done at home (which is amongst the lowest proportion in England). This helps to explain why social distancing has been more effective in some places than others.

The analysis underlines how the nature of local labour markets and job roles will mean different places and groups of people experience the crisis very differently. Whereas this is often seen through an economic lens in terms of the varying abilities of people to continue contributing towards economic activity during the lockdown, this research shows that homeworking (or lack of it) may also be linked to the rate of spread in the disease itself.

50 highest risk authorities (1 = most risky)

Please click here for more details about the health index including data and methods

Download the full dataset

Notes

[1] This is based on a multiple regression model of the proportionate rise in case numbers by local area between April 1st and April 25^th. The key explanatory variable is the proportion of jobs able to be done from home, which is based on the ONS’ analysis of the proportion of broad occupations where people have “ever worked from home” matched to occupational data at local authority level from the Annual Population Survey (APS). Control variables are the number of cases on April 1^st and the size of the population (Population Estimates accessed via Nomis). The model’s adjusted R2 = 0.43 and the coefficient for work from home is -7.