Non-Contributory Pensions and Mental Retirement

Social security
Demographic Economics - Migration
Fiscal Policy - Public and Welfare Economics

Non contributory pensions

Social protection policies for old-age have undergone a significant shift in Latin America. Since 2003, twelve countries of this region have implemented non-contributory pension schemes which, in general, offer a low transfer to elderly individuals who are not entitled to receive any other pension and live in poverty. Transfer generosity, coverage (targeted or universal) and eligibility criteria vary widely across countries, but given the popularity and inherent long-term fiscal commitment of these programs, it is possibly to argue that this is a major change in the strategy to deal with social protection and poverty in old-age. Although the structural pension reform carried-out in the 1990s and 2000s helped to make contributory pension systems more financially sustainable across Latin America, pension coverage rates remained low, with acute differences between rural and urban areas and among income groups.

An important issue is the assessment of the effects of non-contributory pensions on different outcomes related to the well-being of the recipients and their families. Few studies in the region examine the relation between labour supply and retirement decision. Different from the emphasis of those studies, in a recent paper (Novella and Olivera, 2014) we study the potential effects of such pensions on an important health variable of the elderly: cognitive functioning. We use a recent survey conducted in December 2012 among the elderly poor of Peru, which will serve as the baseline to evaluate the program of non-contributory pensions Pension 65.

The motivation is to provide further evidence about the causal effects of retirement on cognitive functioning and the potential effects on public health in the context of a developing country. Our study is in line with a recent strand of economic literature examining the causal effects of retirement on cognitive abilities (Rohwedder and Willis, 2010). It is argued that after retirement the individuals can experiment an acceleration in the decline of their cognitive abilities if they do not maintain an engaged life style, which have been labeled “mental retirement”. This literature is mostly based on industrialized countries where social security coverage is extended and therefore retirement regulation changes can offer a convincing instrument. However, the analysis of the “mental retirement” effect is more challenging in less industrialized economies where social security coverage rates are low, individuals tend to keep working at advanced ages or never retire, credit constraints are large and survey data including cognitive skills questions are scarce. We conjecture that the recent emergence of non-contributory pension programs will induce a significant number of elderly into retirement, and therefore it may have an effect on the acceleration of the decline of their cognitive abilities. Elderly individuals with more cognitive impairments are less autonomous, present a lower quality of life and can represent a burden for the family and a major public health problem in the context of ageing societies.

We use the Survey of Health and Wellbeing of the Elderly (ESBAM) that contains a large set of demographics and objective and subjective health indicators. Additionally, we use administrative data from Pension 65 and exploit the variation of intensity of the program at district level as a source of exogenous variation which might affect retirement. At measuring the relationship between cognitive abilities and retirement we control the influence of different confounders such as schooling, sex, age, household demographics and objective health indicators. This allows us to reduce the potential bias due to omitted variables. Specifically, we use arm-span, which is a better measure than height in old-age population to proxy the nutritional status acquired in childhood that positively affects cognitive ability development. This follows a large body of recent research documenting the importance of accounting for parental inputs and schooling at early ages in the formation of cognitive skills. Moreover, to account for current nutritional status, we use individual levels of hemoglobin. Diagnosed illnesses related to mental disorders are also controlled for.

In order to deal with the potential endogeneity of retirement and cognitive functioning we use an instrumental variables approach. As the main instrument for retirement, we use the number of months since Pension 65 has been operating in the district of the respondent at the time of the survey. We exploit the variation in the timing of implementation of the program across districts and argue that observing a stronger presence of Pension 65 in the district increases the expectation of an individual to receive the transfer. Therefore, the individual can accelerate the decision of retirement, and only through this channel, affect cognition.

Our results show that, relatively to those who continue working, retirement has a statistically significant negative effect on cognition (-0.15 standard deviations in our preferred specification). The size of this effect is about four times larger than the effect of age on cognition. The dimensions of cognitive skills that seem to be most affected are orientation and memory. It is also found that current (haemoglobin) and long-term (arm span) nutritional status are positively associated to the cognitive score, while that having mental disorders is negatively associated to cognition. Being illiterate has a very strong effect on cognition, which lend support to research claiming on the protective effects of education on later-life cognition.

Perhaps, a way to lessen the potential effect of sudden retirement on cognition is not only providing non-contributory pensions to the elderly poor, but also other attached components aimed at maintaining their cognitive functioning or at least retarding its decline. It might be the case that these new non-contributory pension schemes for the elderly should learn from the conditional cash transfer programs that tie the transfer to the accomplishment of some compulsory fulfillments to assure the wellbeing of the recipients.


Novella, Rafael and Javier Olivera (2014) Mental Retirement and Non-Contributory Pensions for the Elderly Poor in Peru. University of Leuven CES Discussion Paper Series 14.05.

Rohwedder, Susann, and Robert J. Willis (2010) Mental retirement. Journal of Economic Perspectives 24(1): 1-20.


Share this