Joseph Stiglitz: preference

Transcript context

…From 1986 until about 1990, you wrote a series of papers with[Raj Kumar] Sah on information architectures, hierarchies versus polyarchies, Type-I versus Type-II errors. Then the 1990s come, and you spend, I think, about seven years working in different roles in Washington, D.C. How did that cause you to revise what you had done with Sah? [laughs] Well, I had never been in a real hierarchy at the time I wrote those papers. Those papers were asking the question, what are the relative merits in decision-making when you have a hierarchy where a decision has to be approved by one person after another, versus what we often think of as the merit of a decentralized economy where you have many, many, many different decision-makers, and you let each of them take their chance. Underneath this was the idea that all humans are fallible. There are going to be some cases of disapproving good projects and some cases of approving bad projects. How do you balance the two? And how do different systems strain out bad projects without straining out good projects? When I came to live in a world of hierarchy, which was what I saw in Washington, I came to appreciate why, in some circumstances, you had that hierarchy. But I think I came to appreciate even more the virtues of decentralization, of what I call polyarchy. I guess I became more of a critic of hierarchy. I saw too many cases where there was too much fallibility built in, and too many good ideas got screened out by the hierarchy, and hierarchies when the guys at the top are not good decision-makers are particularly problematic. Should the World Bank right now be emphasizing climate change, as they seem to be doing? A lot of the poorer nations have complained. They say it’s not their priority. They actually want to use more energy, some of which will be dirty energy. What’s your view on that?…

Stored transcript either side of the excerpt. The highlighted words are the published quote; the surrounding text is unedited source, never generated.