I gave yesterday a lecture on public-private partnerships at a Master for public servants. I gave them material on the literature about formal and conventional infrastructure public private partnerships from economists and business schools, but we spent most of the time talking about their own experience in the local and regional government with projects that involved both the public and the private sectors. It was easy for the students/public servants to come up with examples of projects with public and private inputs.
It was clear from both the examples and the literature that both the public and private sectors had by their nature strengths, weaknesses and common challenges. We concluded that we need high capacity governements at all levels and that strong governments are complementary of good public-private projects.
We focused also on problems of cost underestimation of infrastructure projects, renegotiations and corruption (which are interrelated). Some of this is covered in the work of Chilean economists Engel, Galetovic and Fischer, including their 2015 book on PPPs and their more recent article (with Campos) on the Odebrecht case. Bengt Flyvbjerg has a recent very interesting article on the persistent bias of cost-benefit analysis of infrastructure projects to underestimate costs and overestimate benefits. To counter this trend, he proposes to introduce statistical and institutional measures to de-bias estimations, to punish (with jail if necessary) the more scandalous cases of mis-estimation, and to democratize the evalutaion of public projects along the lines of the work of social psychologist Slovic instead of the technocratic cost-benefit analysis promoted by authors such as Sunstein.
We will need more and better public intervention, complementary of civic participation and consent, to face pandemics and the climate emergency. We need to increase the efficiency and productivity of public projects, and mobilize all the resources of society in a transparent and fair way to face these challenges.
No comments:
Post a Comment