Can pensions impact fertility in Africa?
Traditionally, the issues of social protection and population growth have been studied separately. Where does this idea that there is a link between fertility and the pension system come from?
This is not a new idea. Since the 1950s, sociologists and demographers have highlighted the role of social protection systems, and more specifically pensions, in fertility. By carrying out opinion surveys in the field, they have found that economic reasons appear to be very well placed in the motivations of respondents for having children: people reproduce to keep the family business going, to multiply the family’s income and to provide for their old age. All in all, this phenomenon is quite logical.
The fact remains that these qualitative studies were not able to measure the quantitative impact of this or that measure on fertility: when a government sets up a pension system, does fertility fall by one, two or three children per woman, for example? The study I conducted in Namibia showed that the introduction of a pension system reduced the country’s birth rate by at least one child per woman.
Why did you choose Namibia as a case study?
A few countries in Africa have set up non-contributory pension systems, i.e. they do not depend on the contributions of members, and therefore benefit the entire elderly population. Among them was South Africa which, at the end of apartheid, set up a generous pension scheme, thus becoming a kind of world laboratory for the study of the socio-economic implications of the basic pension. The problem is that there was not enough usable data before this reform to be able to examine the consequences of pensions on fertility in detail.
So, I turned to Namibia – a former South African province – which introduced an equivalent system after independence in 1990. In the mid-1990s, all Namibians over the age of 60 started receiving a kind of old-age minimum equivalent to three times the poverty line ($1 a day at the time), or $90 a month in purchasing power parity. The effect on the birth rate was immediate.
Therefore, since they were anticipating that they would receive this money in the future, women started having fewer children?
Yes, that is what the study shows. In a decade, the birth rate in my sample dropped from almost 6 children per woman to 4. Not all of this is due to the pension system, as we shall see, but the study has shown that the introduction of a universal old-age pension reduces fertility by at least one child per woman. This phenomenon is more visible among older women (over 30), probably because they anticipate their future retirement more.
How can we be sure that other factors did not influence fertility?
This was the difficulty of the study. Many factors can influence a country’s birth rate: schooling, urbanisation, but also the reduction in poverty, access to healthcare, the reduction in infant mortality, or the introduction of other social expenditure, such as family allowances (which tend to work in the opposite direction). In Namibia, pensions constitute 90% of the country’s social expenditure, so there was danger of confusion. To neutralise the other factors, it was necessary to isolate the ‘variable of interest’ (pensions), by constituting two groups of people, more or less affected by the reform.
In this sense, Namibia was an ideal example, as the reform was implemented gradually, so it was easy to measure its effects on different study groups. Before 1992, some ethnic groups already had access to a pension system, but the actual coverage varied from 30 to 80% depending on the region. It was not until the late 1990s that full coverage was achieved. From then on, it was possible to isolate pensions from other phenomena that applied uniformly to all groups. This led to the conclusion that pensions could be attributed to a decline in the number of children of at least one child per woman. And that, overall, the actual amount of the benefit had less of an impact on fertility than the implementation of the system itself.
Today, development aid focuses more on family planning and contraception. Why is this so?
Because it is too often assumed that the number of children does not depend on women’s choices, but on poor pregnancy control and limited access to contraception. In the past, this assumption has been explored with the establishment of family planning and contraceptive programmes in several Asian countries. In the 1970s, an influential study in Bangladesh, for example, found that these measures reduced births by about one child per woman. But today economists are finding it harder to establish the effectiveness of these planning policies on the birth rate, especially in Africa. Every year, billions are spent by OECD governments and large international NGOs to improve access to contraception. Even if this is not their only objective, the impact on the birth rate of African countries is not very visible.