The 2006 analysis: evidence of age discrimination?

The  2006 analysis shows considerable difference between the length of time it take various age groups to find employment. There is a downwards trend as you progress up the age groups. For example, less than half of those aged 16-19 take more than 3 months to find a new job, compared with nearly 7 out of 10 of those aged 55-59.

Old people are more likely to spend longer unemployed than younger people. Arguably, this is evidence of age discrimination in recruitment and selection.

The 2009 analysis: no age discrimination

Compared with 2006, there is more uniformity of treatment.  There is still some variation amongst the age groups, but this is much less pronounced. Those aged 20-64 all took a broadly similar amount of time to find new employment.

There are still some outliers at the extremes of the age range, At these levels, an employer would find it much easier to justify any age discrimination in recruitment and selection. 

The 2012 analysis: increasingly unequal treatment

All age groups are spending longer unemployed. Excluding the extremes of ages, little more than 50% of people find employment within 6 months.  

In addition, the downward trend that can be seen in the 2006 graph is back. Our conclusion from the 2009 data no longer appears substantiated. Each five year age group tends to be more likely to spend longer unemployed than the one before. Age is again a predictor of the time taken to find new work following redundancy.

That said, there are a couple of exceptions: those aged 30-34 and those aged 40-44.

Those aged 30-34 tend to spend a bit less time unemployed than the trend would otherwise suggest. Whilst compared to other age groups, those 40-44 years old spend longer unemployed. Less than 20% of this age group find work within 3 months – the lowest of all age groups.

Another thing to note is that the older age groups were more likely to be out of work for longer following redundancy in 2012 than they were in either 2006 or 2009. 

The data for the 2009, 2012 and 2015 analysis was obtained using a bespoke request to the ONS using variables AGE, DURUN2, REDYL11 and PWTA11. 

The 2015 analysis

This graph is very different to those in previous years. This is because of a change in the way the ONS collect their data. The variable REDYL11 has been replaced with REDYL13. This new variable separates redundancy out into two categories: voluntary and non-voluntary. The graph above represents a sum of both voluntary and involuntary redundancy.

In addition, the population weights are different. The data for 2012 was weighted to the 2011 population estimate. The data for 2015 was weighted to the 2017 population estimate.

Many of the data points were empty as they contained too few individuals to be reliable. This has affected the graph a lot and, as a result, it is difficult to draw any insights out of it. 

The data for the 2015 analysis was obtained using a bespoke request. The data is available here.