Non-Contributory Pensions: desirable and non-expensive

Social security
Fiscal Policy - Public and Welfare Economics

Non-contributory pensions

Implementing a Non-Contributory Pension scheme is desirable and non-expensive: the cases of Colombia and Peru.

Problems regarding the current pensional system in Colombia have been frequently explored by academics and members of the government during the last two decades. Specifically, issues concerning the low coverage and exclusion of the poorest part of the population from the system constitute a problem largely discussed, with no clear solution in the medium run.

Recently, the Journal of International Development published a paper by Javier Olivera and myself, where we studied the ex-ante effects of the implementation of a non-contributory pension (NCP) programme in Colombia and Peru. We simulate the transfer’s potential impact on (i) poverty levels, (ii) inequality, (iii) fiscal cost, (iv) the probability of affiliation to a contributory pension system and (v) the recipients’ labour supply. The potential decline of the labour force’s likelihood of affiliation to contributory pensions is perhaps the most direct behavioural effect one can expect from the implementation of an NCP. For some individuals, the expectation of having an NCP will be a good substitute of pension savings. Two types of transfer programmes are considered: universal and targeted. In the first case, the beneficiaries are all individuals who have reached the retirement legal age and have no pension. In the second case, besides the previous requirements, the transfer is targeted to the poor.

The relevance of an NCP scheme stems from its potential to reduce poverty and inequality, particularly among the elderly. This transfer can possibly become a powerful tool for improving the quality of life of those individuals whose chances of escaping poverty are almost nil. Furthermore, a vast majority of elderly people in rural areas must keep working until very advanced ages (even till death) because they are unable to retire with a secure income stream.

Our results for Colombia and Peru show that the impacts of an NCP (universal or targeted at the poor) are notable for reducing poverty among the elderly, particularly in rural areas. In Peru, old-age rural poverty can be reduced from 48.5% to 23.7% with the introduction of an NCP, while in Colombia, this falls from 36.7% to 24.6%.

The results are modest if one considers national poverty rates, although the effect is higher in rural areas than in urban areas. The impact of the transfer on inequality is rather modest in both countries when one considers the entire population. However, the reductions in inequality are more important and statistically significant among the elderly group living in rural areas. We also observe that there are no sizeable differences between a universal and targeted pension in reducing inequality for rural elderly Colombians. Contrary to this, the targeted pension in Peru is more important to equalise incomes than the universal pension.

Our estimations reveal that the cost of the NCP programmes in Colombia and Peru does not include any substantial crowd-out of contributory pensions and that their cost may be limited when targeted to the poor. In Colombia, the universal programme has an annual cost of 2.6% of the total tax revenues, while the scheme targeted at the poor costs 0.81%. Similarly, in Peru, implementing a universal programme costs 2.98% of total tax revenues, while the targeted programme sums up 1.02%. According to the population projections and under conservative estimates of real GDP growth rate in both countries, these percentages can increase only slightly over the next 25 to 30 years.

As for the possible changes in the behaviour of individuals, we observe that a universal transfer can significantly decrease the probability of affiliation to the contributory pension system in both countries, although the effect is larger in Peru. In Colombia, this programme can reduce the probability of affiliation to contributory systems by about 5%, and in Peru, this reduction is about 8% for men and 18% for women. In contrast, a targeted scheme only slightly reduces this probability, meaning the reduction is less than 1% for both genders in Colombia and less than 2% for Peruvian men. Peruvian women still show a large reduction on the probability of affiliation (10.5%). We also assess the impact of a transfer targeted at the extreme poor and detect very small reductions (less than 0.5%) in this probability for Peruvian men and both genders in Colombia.

Finally, about 2% to 5% and 7% to 11% of the recipients in Colombia and Peru will retire upon receiving the transfer. This effect is positive as one of the goals of NCPs is to ease the path to retirement for the elderly poor.

In summary, our results show that the implementation of the NCP in Colombia and Peru contributes to the reduction of poverty and inequality among the elderly, particularly in rural areas and importantly at limited fiscal costs. In addition, we do not expect a large impact on the probability of affiliation of individuals when the programme is targeted.

Share this