The need for private involvement in infrastructure is unlikely to decline. All of the pressures that contributed to the spread of private infrastructure in the 1990s remain. Infrastructure is still thought to be a key bottleneck to human and economic development, and the governments of the developing countries still lack the financial resources to meet perceived infrastructure needs. The pressure to privatize may decline if the World Bank and other aid providers respond to the current controversy by expanding their lending for government infrastructure projects. But the funding gap is still likely to remain substantial. The World Bank estimates, for example, that the developing countries would have to double their rate of current investment in water and sanitation to meet the millennium development goals.5 Private provision continues to offer the promise that it can help close the funding gap through a combination of improved access to private capital and greater efficiency. The perception that private provision has delivered less than it promised is widespread and understandable. Many advocates of privatization in the 1990s made strong claims about its advantages and contributed to unrealistic expectations. And in retrospect it is clear that we severely underestimated the difficulties of privatization. We often failed to appreciate that the challenge of privatization was not primarily technical, but also fundamentally political. The difficulties arose in three areas: the initial terms of the privatization, the constraints imposed by private capital markets, and the provisions for ongoing regulation.