Covid and the Future of Population Forecasts

It’s hard to know the current population, let alone predict populations accurately for 2050 and beyond.  Yet that’s what many of our clients are required to do for long-term planning.

As we flagged in our recent round-table on the challenges of population forecasting, the government are currently using two different estimates of the ‘current’ population to present the latest Covid vaccination rates. This explains why the English national rate is higher than any of the regions – a mathematically impossibility raised in a recent edition of Radio 4’s “More or Less” podcast.

This is why it’s important to start from a consistent and robust estimate of the current year population by age and gender that users can rely on, at a local level.  Something that CACI achieve using a proven methodology respected by JICPOPS, the Joint Industry Committee for Population Standards, at Postcode Sector level.

We then model it right down to the unit postcode level needed by our many clients that rely on us for an accurate understanding, not just of population numbers, but also their demographic and lifestyle characteristics.

Building on this solid base we project forward nationally as far as 2069 using a consistent set of inputs at a granular geographic level to give credible and affordable ready-made local forecasts. Inevitably the uncertainty of forecasts increases as we look further forward in the crystal ball and we offer custom solutions to our clients seeking to tackle forecasting in areas of greater uncertainty using bespoke inputs.

An uncertain future

But, despite increasing computer power and open data access, it is getting harder to forecast over long-term horizons.  The following are just a few of the challenges faced building forecasts based on today’s uncertainty.

It’s been 10 years since the last Census gave us a solid population base, and many don’t realise we have a long wait until the 2021 Census can feed the latest forecasts. And whilst the ONS reports a great response rate, there may be local nuances resulting from capturing the data during a pandemic.

The jury is still out on where Brexit will level out on the nation’s migration patterns and even climate change could start to impact where people can, or want, to live within current planning horizons – potentially reshaping local populations from a complex mix of local and international movements.

This is before we even think about the unknows from potential changes in planning policies that have moved up the agenda only this week following the Chesham and Amersham by-election.

And then there is the uncertainty from Covid.  This week it was widely reported that UK deaths exceeded births for the first time in 40 years and sadly we know that death rates in specific age and demographic groups have far exceeded long-term patterns, making trend-based forecasting harder.

But will Covid also cause long-term change in local populations in other ways?

Will university cities become a thing of the past, now that lectures have moved online?  Will families seek to support their older relatives closer to home after the challenges seen in the care sector?  Is the ‘race for space’ out of our cities here for the long-term or will people have to return to work in the office despite our survey revealing that most want to return less than 3 days a week.

In short, there’s a lot of uncertainty.  And you can be certain that any of these trends will vary locally and by demographic group.

Our use of mobile data and surveys during Covid has revealed clear insights into the behaviours of consumers during the different stages of the pandemic that can support decision-making into the uncertain future.

I’d love to discuss how we can support you in creating future scenario models using our data. Please get in touch and we can discuss your challenges in more detail.

Paul Langston
Associate Partner | Communities & Government
plangston@caci.co.uk