It is kind of fascinating. How different policy is actually made, compared to political econonmy models. I am reminded a bit of the march and olsen garbage can model, but instead of policy by flight and by oversight, this is policy by pseudo-consensus, generated by people sitting in a room, making comments, randomly interacting, and in the end, people agreeing not because they are happy with the outcome, but out of fatigue, or out of agreement that sufficient process has been conducted.
It has been noted that while economists focus on outcomes, lawyers care about process. Economists have written a bit about preferences over actions (legitimacy) vs preferences over otucomes, but the latter still largely defines the field.
Policy folk like typologies. I worry a bit that much is decided by policy folk who understand politics and process well but have often only a cursory understanding of the scientific and economic details. They like colorful bar charts, and 2 x 2 typologies.
Not so different than whitehouse. A consensus process. Difference is there is no executive, essentially a dictator, of last resort, who may be convinced to make the optimal decision.
This leads to interesting implications for what good policy should look like. Economists would like a global carbon tax (or cap and trade) since that is likely optimal, but given that consensus is required this may no longer work. Even amongst the countries that agreed to Kyoto, half of them failed to comply and many of the ones who did, did so because they had an economic collapse.
This gives further credence to the Bush administration's technologically focused environmental policy.