As promised, here is a blog on the Bolivia chapter of the UNRISD book. It is wonderful, but I limit myself to copying and pasting part of the concluding section.
The Bolivian experience since 1996 clearly illustrates the respective capacity of contributory and non-contributory pension schemes to deliver income security to older people in low income countries. Put simply, it demonstrates that non-contributory social pensions are much better suited to this task. By 2007, the contributory system paid out pensions to around 65,000 older people, of which only 12,000 were paid by the new private scheme. In 2004–5, this contributory system absorbed around 5 per cent GDP in government subsidies, not including substantial indirect subsidies of up to 40 per cent of the annual transition costs through treasury bonds. …. By contrast, the [universal] social pension pays out over 830,000 benefits at a cost of less than 2 per cent of GDP. ….
The most obvious lesson for other developing countries is that non-contributory social pensions are a much more efficient means of meeting the welfare needs of older people than contributory schemes. …. In countries where both exist, government spending on social pensions is usually a fraction of spending on contributory ones. The Bolivian experience also shows that contributory pension funds create powerful vested interests among privileged groups of workers who can resist substantial reforms, even under radical non-elite governments. Only six years after Morales took office was the contributory system meaningfully modified, although many of its essential features remain.
Peter Lloyd-Sherlock and Kepa Artaraz, “Pension Reform in Bolivia: Two Models of Income Security in Old Age”, chapter 9 of Reforming Pensions in Developing and Transition Countries (Palgrave Macmillan, 2014), edited by Katja Hujo, pp. 267, 270.
I came to a similar conclusion in an earlier study of Bolivia:
The cost of the [universal] Bonosol is not low, but benefits go to the entire population of elderly. The annual fiscal cost of reform of the contributory scheme is four times greater, and benefits go to the few, none of whom are poor. Actual costs of reform of the contributory system are more than twice those projected at the beginning of the reform, largely because of fraud, but also because of increased generosity in transferring income from taxpayers in general to the small number of salaried employees (fewer than 12% of the labour force) who participate in the contributory scheme. ….
The Bonosol is a godsend for the poor of Bolivia. …. The World Bank and the Inter-American Development Bank, in a joint report (2004) use strong language to recommend “The Bonosol should be maintained, as it represents a strong redistributive policy with minimum fiscal impact.” For the World Bank, this marks a reversal of earlier views that questioned the worth and the wisdom of the Bonosol, whereas the views of the IMF continue to be quite negative.
Larry Willmore, “Non-contributory Pensions: Bolivia and Antigua in an International Context“, Financiamiento del Desarrollo 167 (United Nations, Santiago, Chile, May 2006), p. 28.
Great minds think alike!
My essay is not as good …. but it is freely accessible.