Think-3 Investment Idea Sharing Platform

"The Numbers Guys" Column in WSJ
September 7, 2005

In the weeks ahead, one of the myriad tasks facing the ongoing Gulf of Mexico recovery effort is tallying a grim number: Katrina's death toll.

With the search for survivors still under way, and weeks of water-pumping ahead just to restore dry land, government agencies have barely begun to count the dead. There have been scattered reports of death counts, almost none from New Orleans, which likely suffered the bulk of the deaths.

If other recent disasters are a guide, it will likely be months before there is a credible number capturing the human toll. There is, so far, no reported official tally of missing persons; the evacuees who would file many such reports are spread over hundreds of miles, and many themselves could, in turn, be reported missing. There are also likely to be many reports that prove to be false; initial counts of people missing in the wake of the World Trade Center's destruction turned out to be inflated by more than 100%.

ABOUT THIS COLUMN 1 The Numbers Guy2 examines numbers and statistics in the news, business, politics and health. Some numbers are flat-out wrong, misleading or biased. Others are valid and useful, helping us to make informed decisions. As the Numbers Guy, I will try to sort through which numbers to trust, question or discard altogether. And I'd like to hear from you at numbersguy@wsj.com3. I'll post and respond to your letters. Also, you can sign up4 to receive email when new columns are published, and you can read more columns at WSJ.com/NumbersGuy5.

The death toll may not matter to the rescue effort, but it will be critical in assessing the government's response to Katrina and in preparing for future disasters. Years from now, it will also be attached to every account of the storm in history textbooks and news articles.

Without much of a functioning city government remaining in New Orleans, it's not even clear yet who will accept reports of people feared dead. Representatives of the Louisiana state police and the state's Office of Homeland Security & Emergency Preparedness didn't know yet who would coordinate gathering the names of missing persons. (Trooper Chavez Cammon said people could call the state agency, at 800-434-8007, and the names would likely be forwarded to the Federal Emergency Management Agency.)

Direct death tolls can't always be measured; for some of history's most fatal man-made horrors, widely accepted casualty figures are derived indirectly from broad population changes -- basically, subtracting population counts after the event from those before it, and adjusting for unrelated births and deaths. (I wrote an earlier column⁶ about similar methods used to quantify the number of Armenians who died in deportation and mass killings during and after World War II.)

But in the U.S. today, it is considered feasible to identify every casualty and report an exact count -- given months of time. And in the meantime, numerical shortcuts for calculating Katrina's death toll, like subtracting the number of evacuees from the city's population, aren't likely to yield accurate results. Widely repeated estimates of the evacuation rate from New Orleans and the number of refugees in other cities are far from precise.

A week ago, New Orleans Mayor Ray Nagin said he thought about 80% of the city's 480,000 residents had obeyed his evacuation order. But it's not clear how the city could have made such a count. I made several inquiries to his press office, relocated to Houston, over the weekend and Tuesday about the number. His spokeswoman, Tami Frazier, said she didn't know how the estimate was arrived at, and that she wasn't sure she could reach Mr. Nagin to ask him. Varying counts placed tens of thousands of people in both the Superdome and New Orleans convention center, and rescuers are finding that many other city residents stayed in their homes.

As for counting evacuees, shelters can produce exact tallies of how many people they have taken in, but it's more difficult to track people who have gone to hotels or people's homes. Baton Rouge arrived at its count of 150,000 evacuees by surveying a few hotels and other sites, and extrapolating those numbers. "It's a rough estimate," Walter Monsour, chief administrative officer of the state capital, told me.

Officialdom abhors a numbers vacuum, and several elected officials have begun to speculate publicly about the death toll. Asked⁷ on NBC's "Today" how many might have died, Mr. Nagin said Monday that "it wouldn't be unreasonable to have 10,000." (On Thursday, he had said in a radio interview that 1,000 had died and 1,000 more were dying every day.) That echoed a statement Friday⁸ by U.S. Senator David Vitter: "My guess is that it will start at 10,000, but that is only a guess."

Educated guessing is an entirely understandable response, and it may help brace the public for the actual number. But that number could be very different.

"There's plenty of speculation. There's a thousand numbers out there, to say the least," Trooper Johnnie Brown, a spokesman for the Louisiana state police, told me on Sunday. I asked if any of the death counts were official numbers. "An official may have said it, but there has been no count," he said. He emphasized that more important work remained to be done: "We don't have anyone who can sit there and be a counter."

By Tuesday, now that the counting has begun -- and a few hundred deaths have been reported -- the state's medical examiner's office will take the lead, with a special FEMA group called the Disaster Mortuary Operational Response Team helping to gather and identify remains. Meanwhile, authorities likely will gather reports of missing persons and eliminate duplicates and false reports. Then the rising body count and diminishing missing-persons count will slowly converge.

It could take two months to gather missing-persons reports from scattered survivors, and three months to recover bodies, estimates Barbara Butcher, director of investigations and mass fatality management for the New York City medical examiner's office, who helped manage the recovery process at the World Trade Center site and is advising authorities coping with Katrina. "At this time, there's still a command-and-control issue," she said Tuesday.

There are more than 23,000 posts on the New Orleans Times-Picayune's online forum⁹ for missing persons, and over 2,000 listings in a separate database¹⁰ for missing-persons reports, but neither is a reliable indicator of the number of dead: Some posts are duplicates and others contain more than one name and there is no way to know how many have already been found.

Some of the news media has shown restraint, either not repeating officials' guesstimates or providing the appropriate context of uncertainty. That wasn't always true in the aftermath of the World Trade Center attack and the Asian tsunami. Those disasters' human tolls weren't fully understood for months, and for both, some early estimates proved to be poorly founded and overstated.

In the first days after the World Trade Center attacks, anonymous officials speculated to the press that 10,000 or more people may have died. Later, the police department began an official count based on missing-persons reports, including often-conflicting numbers from foreign embassies. The count started above 6,000; it remained at about 4,000 two months later. A joint panel representing the police, the medical examiner's office and other agencies put the final official toll¹¹ at 2,749, including those on the planes and in the towers -- in January 2004, more than two years after the attacks.

In last December's tsunami, about 180,000 people died and about 50,000 others are still unaccounted for, according to official reports from the affected countries. But no exact death toll will ever be known, in part because in Indonesia -- which reported more than two-thirds of all deaths -- many bodies were bulldozed into mass graves in the early days of recovery, to prevent the spread of disease. The Indonesian Red Cross nonetheless reported precise figures, but their accuracy was always unclear; the Associated Press reported¹² in March, "At one point, the Indonesian toll jumped by 20,000, only to be corrected downward by 12,000 because an official misheard the number." In April, the Indonesian death toll suddenly plummeted by more than 50,000 when the government removed that many names from its list of the missing, leaving about 164,000 who are thought to have died or gone missing in Indonesia.

Finding an accurate number in New Orleans will, in some ways, be easier than after those previous disasters, and in other ways more difficult. Recovery teams at the World Trade Center found few intact bodies, which won't be the case in New Orleans. Also, Indonesia had less resources and technology to bring to bear than FEMA, whose body-recovery specialists will aid local officials, according to the New Orleans Times-Picayune¹³.

However, New Orleans is still a flooded city, whereas the tsunami waters quickly receded into the ocean. And while most of New York's capabilities and equipment survived Sept. 11, New Orleans is flooded and badly damaged throughout.

Sergeant Nicholas Stahl, of Louisiana's Homeland Security & Emergency Preparedness agency, said Saturday, "We're not worried about counting right now." How soon might the greater emphasis be placed on counting and identifying dead bodies, I asked? His response underscored the challenges facing any such effort: "Probably as soon as we can get the ground crew to build buildings to collect them," he said.