The Associated Press today published an important investigative project by reporter David B. Caruso that details how rising flood insurance premiums across the country will have devastating, long-term impact on many homeowners and communities. Caruso’s report was distributed along with an interactive and AP sidebars from each of the 50 states.
In addition, data shared by AP with member news organizations – such as the New Haven Register and The News Journal (Delaware) – helped them to further localize their coverage.
“This report is a great example of how AP can work with its members and clients to help them produce exclusive, highly local stories that can’t be found elsewhere,” said Brian Carovillano, managing editor for U.S. news. “Going forward, many of our data-driven investigations will include national and state reporting from AP journalists, and our content partners around the country can provide local perspective using data gathered, formatted and distributed by AP.”
Here, National Investigative Editor Rick Pienciak describes how AP tackled the ambitious reporting, which includes data on more than 18,000 communities across the country.
What prompted AP to explore this issue?
David Caruso, who has written a great deal about post-Superstorm Sandy
insurance issues, proposed looking at the big rate hikes in store for
1.1 million participants in the National Flood Insurance Program, 20
percent of all participants. We figured we could perform a nice public
service for a large number of people – in all 50 states.
What were the most striking findings and concerns?
For one, we learned that the public outcry was so strong against the
2012 law to make everyone start paying true-risk premiums (increases as
much as 15-fold), that Congress might push back some of those increases,
many of which had started to take effect in October. In fact, compared
to the usual speed of Congress, a bill was passed by the House and the
Senate earlier this month, spreading out the increases. And President
Barack Obama signed the bill on Friday. (It is worth noting there are
congressional elections in November.)
Instead of paying the full rate immediately, depending on the type of property, those impacted will face annual increases of up to 18 percent or a mandatory 25 percent. In analyzing the data for more than 18,000 communities, we were taken aback by the impact these premium increases will have – even spread out over years – on small, old river towns. The numbers help tell the story that some of these places might very well turn into ghost towns.
What challenges did you face?
One of the biggest challenges was distributing a large amount of data
for so many communities. First, we had to decide what to use and what to
skip from the Federal Emergency Management Agency data because it
wasn’t relevant. And, as anyone who has ever worked with a large,
complicated data set might say, “messy” is an understatement. So David,
AP’s top data guru Troy Thibodeaux and I spent a great
deal of time talking through what we would use, how we’d use it and then
those fellows spent a lot more time “cleaning” the data and getting it
ready for distribution.
One other challenge was the need to distribute the data ahead of time, via a password-protected FTP site, and then prepare our national piece and our state sidebars – one for every state and the District of Columbia – all without knowing exactly when the president would sign the bill to ease the rates of increase. He did so Friday, so we were able to update each story just before they actually hit the wire. We also fielded countless emails and phone calls from member editors and offered them assistance on using the data. Being able to present this package just a couple of days after the president signed the bill is a big accomplishment.
What did AP offer member newspapers to localize this story?
We offered 28 columns of key data for 18,423 separate towns, cities or
unincorporated sections of counties. That is a lot of information.
We provided a unique identifying number for each entity, its name, location by county and state, population, number of policies receiving rate discounts, the number of policyholders facing annual increases of up to 18 percent, the number of policies facing annual increases of 25 percent (that category is generally for vacation homes and businesses).
We also provided the percent of all flood insurance policies in each community facing premium increases. Just from that information, a local reporter can get a really good idea of little towns where large chunks of the citizens are going to face hard times until Congress comes up with a long-term fix.
We provided data state by state, and by type of structure (a business, residential home or 2-to-4-family building). And then there were columns for claim history, numbers of active policies in each community, the total annual premium in a community, total payments since the community joined the federal insurance program and total number and dollar amounts of claims paid out to each respective community. Local reporters could really dig in deep and write their own stories about their town or city, regardless of size.
