Raise a glass

FEW issues have generated more heat at this week's third world water forum in Kyoto than the privatisation of water supplies. Only dam-building arouses more hostility among non-governmental organisations (NGOs). Water, in their view, is too precious and too important to be put into private hands. Private companies, runs the argument, will care about profits, not about the impact of water on human health and disease.

A new study* by three economists, two based in Argentina and one at the Haas School in Berkeley, California, suggests that the NGOs are wrong. Examining Argentina's privatisation campaign in the 1990s, one of the world's biggest and one of the few to involve extensive privatisation of local water services, they find that privatisation may actually bring benefits for the health of young children, the group most vulnerable to disease.

The study takes advantage of the fact that water services in Argentina are the responsibility of local governments, and that only 30% or so of municipalities chose to privatise them between 1991 and 1999. That creates a control group for comparing child mortality.

Water privatisation in Argentina certainly brought increases in productivity and profitability. The largest privatisation involved the transfer to Aguas Argentinas, a consortium led by Lyonnaise des Eaux, a French company, of OSN, a federally owned entity in Buenos Aires. At the end of the first year, prices for both water use and connection were lower than they had been at the start. Non-payment of bills had been high; by cutting customers off after three unpaid bills, the company got 90% of people to pay. The number of employees, whose average age was over 50, was cut by almost half. In its second year, Aguas Argentinas was highly profitable.

In the decade before privatisation, OSN had invested too little to replace depreciating assets. In the late 1980s, the proportion of the population receiving water had been falling and summer shortages frequent. From $25m a year in the decade before privatisation, the company's investment rose to about $200m a year in 1993-2000. Connections to the water and sewerage networks rose, especially among poorer households: most richer households and families in the city centre were already hooked up. Besides, the government set targets for connecting the poor, who have also been largely protected from subsequent disconnections.

Other water privatisations in the country seem to have achieved broadly similar results. Among municipalities in Argentina that privatised in 1990-96, the number of households connected to the network rose by 4.2 percentage points more than in districts that remained in public hands.

Before privatisation really got under way, in 1995, child mortality rates were falling at much the same pace in municipalities that eventually privatised and those that did not. After 1995, the fall accelerated in privatising municipalities (see chart). The fall was concentrated in deaths from infectious and parasitic diseases, the sort most likely to be affected by water quality and availability. Deaths from other causes did not decline.

These effects were much stronger in municipalities with high levels of poverty than in those that were better off. Overall, childhood mortality fell by 8% in areas where water services were privatised, and by 24% in the poorest areas. The NGOs in Kyoto campaigning against privatisation are therefore likely to be condemning more children, not fewer, to death from waterborne diseases.

* “Water for Life: The Impact of the Privatisation of Water Services on Child Mortality”. By Sebastian Galiani, Paul Gertler and Ernesto Schargrodsky.