That's nacho cheese

This graph is floating around the interwebs (click on image for a larger version):



It's from Daniel Wilson of the SF Fed who wrote a short piece entitled, "Are fiscal stimulus funds going to the "right" states?" He defines "right" as "have the funds been allocated in a way that maximizes their potential impact on national economic growth?"

Wilson says that the answer is basically yes.

My first reaction was to wonder what exactly is going on in Connecticut?

My second was to look at the fine print. Here is what I found.

(A) The stimulus is $787 billion. The graph covers $144 billion in the Fiscal Stabilization Fund and the Fiscal Relief Fund.

(B) The data in the graph are projected state deficits and estimated state allocations from the two funds. In other words, neither variable has actually happened yet! Both estimates come from the Center on Budget and Policy Priorities

(C) 12 states are excluded from the graph.

(D) The regression line is from a population weighted least squares regression and all the piece says is that "The figure clearly shows a strong positive correlation between a state’s degree of fiscal strain and the amount of federal stimulus funds it is expected to receive", with no indication if "strong" means significant.

My third reaction was to remember Gavin Wright's classic 1974 RESTAT paper on the political economy of new deal spending where he claimed that a "political" model ex-plains between 58.7% and 79.6% of the variance in per capita spending over the whole period !"

If Team Obama can make the overall stimulus as economically well targeted as the above graph suggests one particular chunk of it might turn out to be, that would be progress indeed.

Hat tips to Bob Tollison and Menzie Chinn