Unemployment by Constituency: changes in methodology, 2016

The House of Commons Library publishes a monthly briefing paper, Unemployment by Constituency. This provides the latest data on the monthly count of people claiming unemployment benefits in each parliamentary constituency.

The Library calculates a claimant rate for constituencies which shows the number of claimants as a proportion of the estimated economically active population aged 16-64. From the January 2016 edition of Unemployment by Constituency, we have revised the methodology used to calculate the claimant rates to take account of the latest constituency population estimates for mid-2014 and to reflect recent trends in economic activity at the national and regional level.

The attached table shows how the change in methodology affects claimant rates in May 2010 and in November 2015.

What changes have we made? Population growth

The denominator for the Library claimant rate is the economically active population aged 16-64. Before January 2016, the economically active population was estimated using 2011 Census data for parliamentary constituencies, scaled to reflect population growth in each constituency up to mid-2013.

From January 2016, we have revised the denominators to reflect population growth in each constituency up to mid-2014, following the publication of new population estimates by the ONS, National Records of Scotland and the Northern Ireland Statistics and Research Agency.

Changes in economic activity

The total UK population aged 16-64 has increased since the Census but the proportion of the population who are economically active has also increased. Therefore we have revised the denominators to reflect growth in economic activity rates since 2010.

We make the assumption that any change in economic activity at the regional level is uniformly spread across all constituencies (although clearly we would expect a more varied picture in reality). Data on trends in economic activity at the country and regional level is available from the ONS Annual Population Survey. We have scaled our constituency denominators for May 2010 to July 2014 so they sum to the national and regional figures from the Annual Population Survey.[1]

The effect of this change in methodology is relatively minor for more recent periods from July 2014 onwards. However it has a greater impact on claimant rates in 2010 and 2011 – in most cases it means there will be a slight upward revision to claimant rates in earlier periods. This is because the Census gives higher figures for economic activity in 2011 than does the APS and we have scaled the denominators downwards to match the APS figures.[2] A smaller denominator means a higher claimant rate.

Other sources

The Library claimant rate differs from that calculated for constituencies by the Office for National Statistics (ONS) and used in Written Answers. The ONS rate shows claimants as a proportion of the total population aged 16-64, rather than the economically active population. Someone is economically active if he or she is either in work or is looking for and available to start work. Consequently the Library rate is higher than the ONS rate, since the Library rate uses a smaller denominator.

The claimant rate should not be confused with the unemployment rate. Around twice as many people are unemployed in the UK as are claiming unemployment benefits (i.e. people claiming Jobseeker’s Allowance or who are not in work and claiming Universal Credit). Therefore the unemployment rate for a constituency will be higher than the claimant rate.

Estimates for constituency unemployment rates are available from the APS, but small numbers of survey respondents mean constituency figures are much less reliable than estimates at the regional and national level. Estimates are generally not robust enough to allow for analysis of changes over time or comparisons with other areas.

[1] Quarterly data on economic activity is also available from the ONS Labour Force Survey. We have used the Annual Population Survey in preference to the Labour Force Survey because it has a larger survey sample and so gives estimates that are more robust. ONS does publish constituency figures for economic activity based on the annual population survey, but small numbers of survey respondents mean constituency figures are much less reliable than estimates at the national or regional level.

[2] There are a number of reasons for this difference between the Census figures and estimates from the APS and Labour Force Survey. These are discussed in a 2012 article from ONS, A comparison of the 2011 Census and the Labour Force Survey (LFS) labour market indicators