On Friday morning, October 15, Kestin again ventured into Miami-Dade’s public housing to gather a few more details to bolster the story. She then hurried back to the office to join O’Matz, who had begun to write an article for Sunday’s paper. They worked late into the evening shaping details into a strong narrative. They discussed which examples best illustrated the emerging pattern of fraud. It was difficult to decide which of dozens of details and sources to leave out. Due to space constraints, they could not tell all they knew. Each had notebooks full of interviews, in addition to all the other data and documents they had gathered in nearly a month of reporting. “One of the hardest things,” Kestin says, “is maintaining a focus.” She continues:

You’ve got this massive amount of information... How are you going to take that and get the biggest bang for your buck [and] write a clear, high-impact story that boils [it] down?

Sources. The strongest stories had come from the people who had admitted to committing fraud themselves, but all of them had refused to give their names. That posed a problem for the team. All agreed that it was best not to quote anonymous sources directly; doing so, they thought, undermined the newspaper’s credibility. Still, Kestin felt comfortable using anonymous sources to make authoritative statements about general patterns of aid distribution in the projects, so long as their stories were sufficiently consistent. She was stunned at some of the creative uses of disaster relief funds: People had used them to purchase new cars and jewelry. One woman even admitted to Perez, without identifying herself, that she had paid for her wedding with FEMA money.

Kestin and O’Matz had several attributable quotes, but these were mostly from sources who had insisted that Frances had damaged their homes and had shown the reporters their rundown apartments as proof. So the two reporters contrasted residents’ descriptions of the storm’s impact with their own observations of the slight damage visible. They supported their impressions with accounts from building owners and managers, who consistently denied knowledge of damage at properties where many tenants had received FEMA money. The juxtaposition created a tone of skepticism about whether those who had received aid in Miami-Dade were legitimate disaster victims.

When he reviewed the story late that evening, Demma was amazed at the amount of information his new team had uncovered. “They went into some of the worst projects—crime-riddled, drug-riddled, dangerous areas, and came out with quotes about how FEMA worked in those projects,” he recalls. “I’ve been doing this going on 43 years and I’m always amazed [that]... people will talk to us, and what they’ll tell us.” From that standpoint, the story seemed solid.

Risking criticism. But there was a problem. The team worried that reporting widespread fraud in poor, heavily black communities could open the Sun-Sentinel to accusations of insensitivity or racism. They knew that this was an especially sensitive point for Managing Editor Rosenhause, who prided herself on a diverse newsroom staff; the paper even had a team of reporters devoted specifically to covering racial issues. “It was on our minds from the beginning,” Kestin says. But because FEMA offered assistance only for uninsured losses, she continues, “you just don’t see the money coming into [affluent] neighborhoods, and when it does, the amounts are tiny compared to what was going to the poorer neighborhoods.” O’Matz summarizes: “We have a great deal of sympathy for people that are really struggling... [But] fraud is fraud, and so that's what we were pointing out.”

They agreed that it was unnecessary to identify in writing the race of their sources. “It would be very easy to turn this into a race issue,” Demma reflects. “And the truth of the matter was, it was a system issue.” They hoped readers would see it that way too, but they braced themselves for criticism. By the time the article ran on Sunday October 17, one week after the first, FEMA had dispensed $23.6 million to 10,568 residents of Miami-Dade County. ^[30]