Thinking in Systems
On Cennydd's recommendation, I read Thinking in Systems over a couple of work commutes. There's a lot of interesting stuff in there, particularly this paragraph towards the end:
For example, in 1986 new federal legislation required U.S. companies to report all chemical emissions from each of their plants. Through the Freedom of Information Act (from a systems point of view one of the most important laws in the nation), that information became a matter of public record. In July 1988 the first data on chemical emissions became available. The reported emissions were not illegal, but they didn't look very good when they were published in local papers by enterprising reporters, who had a tendency to make lists of the top ten local polluters. That's all that happened. There were no lawsuits, no required reductions, no fines, no penalties. But within two years chemical emissions nationwide (at least as reported, and presumably also in fact) had decreased by 40 percent. Some companies were launching policies to bring their emissions down by 90 percent, just because of the release of previously sequestered information.
Food for thought.