Weathering the Storm
Restricted census data reveals which businesses are best able to survive a hurricane.
Besides being lucky, the best way local retail and tourist businesses can survive a hurricane is to be big, smartly financed and have lots of revenue for its size.
Those are the conclusions Associate Professor of Economics Emek Basker reached after analyzing detailed, firm-level data from the U.S. Census Bureau for the four Mississippi counties most damaged when Hurricane Katrina struck the state Aug. 29, 2005. She posted her paper, written with co-author Javier Miranda of the Census Bureau, Oct. 23, 2014, on the Social Science Research Network.
The census data used for the study — available only at a handful of restricted-access Research Data Centers, one of which will open in Ellis Library in the next year — allowed Basker to track which businesses either remained open or reopened after the storm and which added or lost employees. The data also contained business locations, which, when coupled with detailed maps from the Federal Emergency Management Agency, told her which businesses Katrina damaged.
Basker looked at retail stores, restaurants, hotels and casinos. Location is key for such businesses, which serve locals and tourists. She chose these because it would be easier to isolate Katrina’s effects compared to businesses in a global marketplace.
Basker found two straightforward keys to business survival. In percentage terms, storm-damaged businesses were 30 points more likely to close than undamaged ones (although the economic disruption of the storm strained even undamaged businesses), and businesses that were more productive (those that had more revenue given the number of employees) were more likely to reopen than less productive ones.
But she also spotted less predictable trends.
For instance, bigger is better. Basker found that, in areas of Mississippi unaffected by the storm, large firms closed less frequently than small ones, independent of how much business they did. But that advantage tripled or quadrupled among businesses damaged by Katrina.
The type of financing used also matters. Firms were more likely to close if they relied on credit cards, rather than traditional loans, to pay for business expansion or capital improvements.
Basker could not say definitively why smaller businesses fared worse but speculated that difficulty accessing traditional credit, limited collateral to borrow against and greater risk aversion might all play a role.
Basker’s paper “could not have been done without access to the confidential data maintained by the Census Bureau,” she says. Mizzou researchers used to drive as far as Chicago or Minneapolis to reach the nearest data center. “Most of my current research uses confidential census data,” Basker says, “and there is no other way for me to answer the sorts of questions I’m interested in.”