Pension reforms have been central to all Memoranda of Understanding signed in Greece during the recent crisis. Although the reforms aimed at a radical change of the pension system, which experienced serious problems for decades before the crisis hit, structural reforms did not really bite during the crisis years. Thus, while the first pension reform following the loan agreement (Act 3863/10) increased retirement age for the next generations, if anything it largely incentivized early retirement during the crisis, especially among women, in an effort to shrink the public sector. It soon became clear that the targeting of the 2010 reform despite addressing pension sustainability, addressed structural imbalances in the long-term.

What really seems to have hit pensioners during the crisis years was pension cuts. In an effort to reduce public expenditure, pensions were reduced on ten separate occasions between May 2010 and mid 2013 with further smaller cuts in 2014 and 2015. Low pensions were hit predominately by the abolition of the 13^th and 14^th pension (holiday bonuses) in 2013. All in all, between May 2010 and September 2015 gross pensions suffered significant reductions, ranging from 15% for low pensions to almost 50% for higher ones. It seemed that the only criterion used in cutting pensions was their level, rather than a thorough justification of those cuts on the basis of age or contribution years.

So how has the crisis affected the consumption of those that were about to retire in the period of the financial crisis?

Household expenditure and consumption during transition to retirement have been the subject of long debate in the literature. Economic theory suggests that savings are used to smooth consumption over the life cycle, yet empirical evidence in the past three decades has been conflicting and inconclusive. Evidence from other countries points towards a “retirement consumption puzzle”, suggesting that household expenditure drops as individuals transition into retirement. Yet, so far the literature on income smoothing treats retirement as a predictable change in one’s life and little is known about what happens to the consumption of those approaching the retirement age during an economic crisis period.

We study the impact of retirement of couples on household expenditure post 2008 in Greece, considering both own and spousal retirement status. More specifically, we examine whether there are changes in total household expenditure and expenditure sub-categories in Greece at the Early Retirement Age, which is a crucial policy threshold. In general, retirement age-related thresholds are considered to be associated with important behavioural and lifestyle changes (Fitzpatrick and Moore, 2018).

Using detailed household data covering the period 2009-2016 in Greece from the Hellenic Statistical Authority (ELSTAT) we find strong evidence that total household expenditure drops significantly when the husband retires, but not when the wife does. This may be either because the wife is often the second earner in the household or because she is usually younger than the husband, so by the time she reaches retirement age, the household has adjusted to income changes. Our findings also suggest that the drop in total expenditure is driven mainly by a reduction in clothing, transport and communication supporting previous claims that individuals reduce their spending on work-related expenses. However, we find further reductions in certain subcategories such as recreation and culture suggesting that both men and women spend less on things that one would do in their leisure time. This may be related to the crisis effect. Indeed, total household expenditure reduces even further for retirees post 2014, suggesting that financial crises have an impact on retirees and their household expenditure. The findings have policy implications concerning pension reforms that are planned during financial crises, as these have an impact not only on retirees themselves but also the household expenditure in total. However, our research on this topic continues with more and richer datasets in order to have an even clearly picture regarding this link during turbulent times.

A research seminar on the topic took place on 10^th December 2019 at the LSE, organised by the Hellenic Observatory. For more information please visit the event page. A full working paper can be found here.

Note: This article gives the views of the authors, not the position of Greece@LSE, the Hellenic Observatory or the London School of Economics.