Like many industries and practices, the field of emergency management is on the cusp of significant change. Advances in technology and data science can now deliver highly precise and actionable insights that were previously unattainable — insights that enable smarter, more surgical approaches to revitalizing regions hard hit by natural disasters.
“By overlaying flood data, hurricane data, or fire data over a detailed economic activity map, we can say, ‘Here is our economic risk profile,’” said Joel Thomas, CEO of SPIN Global LLC, a Washington-based emergency management consulting firm. “Then we can understand the populations we’re trying to serve in concrete and detailed terms, not just generalities, and have a more complete picture of exactly what needs to be done to get a community back on its feet after a disaster.”
This sort of data-driven approach to disaster recovery is being practiced in a few leading-edge jurisdictions, though it remains more the exception than the rule. In most places, economic impact data still plays only a limited role in disaster preparedness and recovery. Emergency planners, for example, use meteorological data to plot the severity of disasters and the scale of response activities to follow. And geospatial data provides important situational awareness of a disaster’s impact on a region.
These datasets help decide where to focus efforts before and after a disaster to protect lives and property. But this data alone is not sufficient to effectively guide disaster mitigation, preparedness, and recovery efforts in support of the affected regional economy.
“When it comes to disaster mitigation or disaster preparedness today, the emergency management community tends to be reactive,” said Tim Nealis, Vice President of Government Solutions at Dun & Bradstreet. “When big events happen, the big effort we typically see is figuring out where to place the dollars, especially in the recovery phase of a disaster, to get the best return so businesses can get back online and people back to work.”
This is a critical point, especially when a single major hurricane can run up tens of billions of dollars in recovery costs. Most state and local governments have little data to assist them when deciding where and how to allocate resources. Instead, they are forced to be reactive, often influenced by political or anecdotal considerations. But some governments are taking a more data-driven approach. The governments of the District of Columbia, Prince William County in Virginia, and the state of North Carolina, for example, have developed highly detailed baseline assessments and business risk profiles of their regions’ economic ecosystems. These assessments inform planners of exactly where economic activity in their regions is strongest and weakest, what types of businesses operate in their regions, how many people they employ, how those businesses interrelate, and, ultimately, where recovery dollars would do the most good in revitalizing the local economy.
Understanding early how businesses rely upon each other can be especially important for emergency responders. Getting a big factory up and running after a natural disaster is obviously important; and figuring out the locations and status of that factory’s critical parts suppliers is a challenge. This is where good, insightful data can be helpful, says Dun & Bradstreet’s Nealis. “Good economic impact data provides deep visibility into those important connections between businesses. If I’m a small business, I have a mortgage, I have creditors, I have banks that I work with, I have suppliers that I buy from. We can see all of that activity.”
With this kind information in hand, emergency managers can better decide where and how to prioritize resources for infrastructure repairs, construction projects, and business assistance for companies that have a disproportionate impact on their local economy. They can better tailor recovery policies and programs to the actual needs of the region. And they can more easily calculate and justify funding requests needed to get a hard-hit region back in business. Ultimately, this translates into better-informed decisions, fewer wasted resources, faster economic recoveries, and measurable results.
Michael Sprayberry, Deputy Homeland Security Advisor for the state of North Carolina and Director of North Carolina Emergency Management, said having detailed economic impact data on his state was critical in the aftermath of Hurricane Matthew, which pummeled the East Coast in 2016. “We found that by utilizing Dun and Bradstreet to perform an Economic Impact Analysis of Hurricane Matthew, we were able to possess factual, data-driven information to assist the recovery team in making assessments and operational decisions for resources and funding,” Sprayberry said. “Otherwise, we would have been restricted to making decisions based on what we ‘thought’ we knew, anecdotally. We were able to determine exactly what impacts Hurricane Matthew had upon North Carolina’s business sectors. The analysis helped the state to craft a request for additional recovery funding, as we could accurately portray data sets, such as employee contractions, changes in sales by towns, etc. Especially valuable was our ability to see detailed effects on the agricultural industry and demonstrated increase in business viability risk.”
The newest frontier in data-driven emergency management, Nealis says, is employing predictive analytics. Analytical models, based on data from past disasters such as Hurricane Matthew, are now capable of predicting with great accuracy which businesses are best positioned to bounce back after a major disaster — critical information when deciding where to dedicate limited recovery dollars after a disaster.
“This is the next stage that we're getting into now — examining what has happened in previous disasters,” Nealis said. “Here's what the area looked like prior to the disaster, here’s how it looked when the disaster occurred, and then what it looked like a year down the road, two years, three years, four years down the road. If we overlay where actual recovery dollars were spent, we can see what impact those funds had to which industries and over what time period. Our models have been pretty accurate in predicting where the successes occur and where to fund. The emergency managers we have shown this to really appreciate how this predictive aspect of the data can help them become far more precise in how they allocate funds after a disaster."
Thomas of SPIN Global predicts big changes ahead as emergency managers employ more and more data analytics into their planning. “This is going to be a game-changing moment for emergency management,” he said. “Having this kind of data available means they can consider return-on-investment (ROI) types of decisions. What’s the ROI on recovery activities, on grant dollars, on loan programs, on our assistance programs? What’s been the result in community preparedness, in economic development, in jobs sustained, and jobs gained? That will be a watershed moment when we can stand on the mountaintop and say, ‘We can now be far smarter and more effective in managing disasters.’”