Doing The Math on Trust in Government

I've always thought there were some real problems with the predominance of polls showing that, for example, two-thirds of Americans are annoyed at the federal government. For one things, satisfied/dissatisfied numbers don't actually give anyone a sense of what the public would like the government to do differently. And as John Sides points out at the Monkey Cage, those numbers don't last: trends in feelings about government are show a cyclical rise and fall that, if it's linked to anything it's the economy, rather than a broad and consistent rise or slide. And I'm not sure what that tells government observers, other than that the American public in a loose way holds the government responsible for the economy's performance. That data is a good hedge against drawing misguided conclusions. But it doesn't really demonstrate a way forward in terms of actual governance, and actual management of agencies and programs.