This story has been covered already by a few blogs, but not here, and none seems to have presented the graphic shown below.
Jowie Chen, a doctoral student in political science at Stanford, obtained the records of 2.6 million applications for FEMA disaster assistance in Florida in 2004. (This happened not because the Bush administration made them available voluntarily, but because four Florida newspapers won a Freedom of Information Act suit.) Chen then mapped these applications and the subsequent FEMA grants into counties and precincts, added data on actual weather conditions in these locations, their demographic characteristics, and finally local voting patterns in recent Federal elections. The paper can be found here. The following map shows part of his results.
</DIV>
The above map shows the 26 Florida counties whose residents were eligible to apply for individual FEMA aid after Hurricane Charley. Dixie and Levy counties were declared eligible by Bush, even though both counties were far away from the storm and had calm weather. A row of counties to their south were closer to Charley’s path but not declared eligible. Also excluded were the Democratic-leaning Tampa and St. Petersburg, which were much closer to the storm's path.
Other key findings:
- Solidly Republican precincts received 19% and 42% more aid than moderate and Democratic precincts, respectively, on a per-application basis. This takes into account different weather, different propensities to apply for aid, income and other demographic factors.
- Variations in storm conditions accounted for only 30% of the variation in FEMA payments -- more variation (about 50%) was due to political or demographic factors.
- FEMA aid was effective in boosting turn-out and the Bush percentage of the vote when it was disbursed in Republican-leaning counties. In such counties, FEMA aid increased the probability that registered voter would actually vote, and that they would vote for Bush. In swing counties the effect of FEMA aid was significantly weaker, while in Democratic-leaning counties, FEMA grants slightly boosted turnout but had no effect on the vote for Bush.
- The presidential authority to declare counties eligible for FEMA's Individuals and Households Program is directly connected to the pattern of FEMA grants. While all 67 Florida counties were declared disaster counties under Hurricane Charley, President Bush declared only 26 of the counties eligible for individual disaster aid. (In the remaining 41 counties, only local governments were eligible to request public assistance.)
The final twist to the story is ironic but, unfortunately, not surprising: none of the Florida newspapers that filed the original lawsuit seems to have published anything on this (at any rate, I could not find anything).