1. Introduction
The level of income is one of the most important factors that impact the welfare of an
individual, and possibly the most important economic factor. Several events may trigger
considerable income drops, one being exit from labour force. Labour income accounts for a
large percentage of the income of all persons. According to Heinrich (2000), in Portugal,
that percentage was the largest one of the fourteen European Union countries that he
studied (59% in 1996). Therefore, as exit from the labour force produces a substitution for
the main source of income, someone interested in analysing income dynamics should
definitely pay attention to what happens around such an event.
The demographic ageing of the economies has put a new emphasis on issues related with
the welfare of the elderly. In this paper, we are concerned with the evolution of the income
of the elderly. Following the argument in the previous paragraph, we should then look at
what happens when an elderly leaves the labour force. But in the late part of life, that is
almost the same as looking at what happens when someone retires.
Hence, this paper addresses the question “is the transition into retirement associated with an
increase in the probability of becoming poor?”. The unit of analysis is the individual that
retires: he/she is not necessarily an old person, but the income effect of retirement is
certainly going to have consequences on how he/she is going to spend his/her old age.
Since the most dramatic change in income is the one that leads someone into poverty, we
focus specifically on poverty dynamics rather than on general income mobility.
A large literature exists about poverty in old age, mainly based on cross sectional survey
data, but usually those studies are not able to study directly the effect of retirement on
income as they do not observe the workers who do retire before and after their retirement.
The knowledge of this phenomenon is, however, of crucial relevance given the growing
number of elderly people, the trend towards earlier retirement, and continuing relatively
high poverty rates among the elderly. This is more so when we analyse the Portuguese
situation. Portugal has specificities that turn it into a natural case study: it combines high
poverty rates among the elderly - using both cross-sectional and longitudinal measures -
and high inequality among the elderly, with generous substitution rates concerning