Media Briefings


  • Published Date: April 2013

Retirement doesn’t mean the end of paid work, according to a study by Ricky Kanabarpresented at the Royal Economics Society’s 2013 annual conference. His research analyses data from the English Longitudinal Study of Ageing (ELSA) and findsthat paid work after retirement – so-called ‘unretirement’ – is common among British men, particularly those who were higher earners in career occupations.


The study followed 592 retired men, tracking any changes in their reported working status between 2002 and 2009. During this time, over 5% switched back into paid work, working on average 20 hours per week and earning on average £470 per month in their ‘unretirement job’.


Certain factors were associated with an increased likelihood of working in retirement: having at least an A-level qualification; having a financial planning horizon of one year or more; experiencing a financial shock; having only been in retirement for a few years;and having a partner in paid work. The last point suggests that men may prefer to retire at the same point as their partner to enjoy a shared retirement.


The study also uses information about individuals’ work history prior to initial retirement. It finds that unretirement was more likely among higher earners who had career occupations – the average pre-retirement salary for those who unretired was around £50,000.


The author argues that, combining this with the evidence that higher educated groups and those who have longer-term financial planning horizons are more likely to ‘unretire’, suggests that unretirement may only be feasible for certain types of retired British men.


He adds that these findings are relevant to the government’s aim to extend people’s working lives in an ageing society, noting:


My study shows that unretirement is prevalent among retired British men. Clearly this is a group of individuals who have a vast number of years’ experience in the labour market, are likely to be highly qualified and can thus contribute significantly to the British economy.


Moreover, given the country’s ageing population and increasing life expectancy and the fact that until only recently there had been no rise in the age at which individuals were eligible to claim their state pension, then government policy should look to provide incentives for unretirement behaviour.’




A significant increase in life expectancy over the past 40 years has meant that many governments are faced with ageing populations. It has only been in the past decade that governments have legislated increases in the state pension age. The 2011 Census figures showed that one in six people are aged 65 and over, and the Office for National Statistics reports average British male life expectancy at 78 years of age, so there is a sizeable pool of workers with untapped economic capacity.


The focus of this study is to determine the prevalence of unretirement behaviour and the economic characteristics of those who undertake such behaviour.


Using the ELSA, the study tracks 592 British retired men who were at least 50 years of age in 2002 for the subsequent seven years following (that is, until 2009) to determine the extent of unretirement. It finds that over 5% of the sample exhibited unretirement behaviour. It then modelled the duration that individuals had spent in retirement as a function of their economic and socio-demographic characteristics.


It finds that the likelihood of observing unretirement was higher if a British retired male had:


A wife in paid employment
At least an A-level
Been in retirement less than eight years
Had a financial planning horizon of at least one year


These are important findings. They suggest that unretirement is associated with certain types of older males and may not be an option for all retirees, for example, employers may look for individuals who are highly qualified. It is also interesting to note that having a partner in paid work increases the likelihood of a return to work, which may suggest couples have shared leisure preferences during retirement.


To determine the types of occupations those individuals who unretired held prior to retirement, the study links its sample to a one-off ELSA life history dataset. In support of the earlier observations, those who unretired tended to hold ‘career jobs’ with an average annual salary of £50,000.


Thus, it is likely these individuals had some taste for work and after a short period in retirement felt they wanted to return to paid work. Indeed those who did worked on average 20 hours per week in their unretirement job, akin to a partial retirement job,and earned on average £470 per month.





Notes for editors:


‘Unretirement in England: An Empirical Perspective’ by Ricky Kanabar.




Ricky Kanabar

Department of Economics & Related Studies

University of York


RES media consultant Romesh Vaitilingam:

+44 (0) 7768 661095