All economists are interested in the unintended consequences of government policy. When the Bush Administration passed pro-ethanol policies, some Princeton researchers (Searchinger et. al) argued that such policies would have severe consequences.
Now Tom Darlington has written his critique of the critique. Darlington's Ethanol Report
As an outsider, I find this interesting and important but I'm struck by how complicated this is. There are a huge number of parameters that need to be well estimated in order to generate a "best guess" concerning the likely size of the unintended consequences of ethanol policy. Put a standard error on each of these parameters and the confidence interval for the final prediction will be quite large.
When our best models yield such loose bounds on the likely consequences, are we useful people for the politicians?