11.522: UIS Research Seminar (Fall 2014) - Discussion notes

Monday, October 20, 2014, 7-9 PM

Population Estimation and FEMA Flood Maps

Discussion Leader: Jonathan Goldman

Background:

The National Flood Insurance Program and FEMA Flood Mapping

Since 1968, the United States Federal Emergency Management Agency (FEMA) has run the National Flood Insurance Program (NFIP), which offers insurance to cover losses from floods. As part of the NFIP, FEMA creates Flood Insurance Rate Maps (FIRMs), which show flood zones according to how often they are expected to be inundated. The Special Flood Hazard Area (SFHA) is the area expected to be flooded at least once every 100 years; all homes in the SFHA that have a federally backed mortgage or receive a federal subsidy are required to maintain flood insurance.

Although the total annual premiums for NFIP policies are intended to be sufficient to cover all losses in an average year – i.e. the NFIP is supposed to be self-funding – this has not worked in practice, and the NFIP currently has a deficit of over $25 billion. This is largely because (a) buildings that existed before the area was mapped for flood risk (“pre-FIRM” properties) received subsidized policies; (b) policyholders were able to “grandfather” their premiums, so that even if subsequent studies showed an increased flood risk, their premiums would not rise; and (c) the flood mapping process relies on historical data, which tends to underestimate risk during this current period where our expectations about the frequency and severity of catastrophic events have been increasing.

NFIP Reforms, 2012-2014

In order to better incentivize hazard mitigation and ensure the NFIP’s future financial sustainability, Congress has reformed the program numerous times. Most recently, the Biggert-Waters Act of 2012 was an attempt to raise all NFIP premiums so that they reflected the full flood risk; however, the Homeowner Flood Insurance Affordability Act of 2014 then scaled back and delayed the impact that Biggert-Waters would have had. The main results of the 2012 and 2014 reforms are as follows:

1. Pre-FIRM subsidized premiums will increase by 18% per year* until they reach the full-risk rates.
2. Grandfathered premiums (i.e. those for properties whose flood risk zone increases) will also increase by 18% per year until they reach the full-risk rates. (... and 25% if they are not for primary residences.)
3. Policies for buildings newly mapped into the SFHA will begin at a reduced rate, but will increase by 18% per year* until they reach the full-risk rates.
4. Deductible limits for residential polices will increase from $5,000 to $10,000.
5. All policies for primary residences will include an annual $25 surcharge; all others policies will include a $250 surcharge.

While these reforms do set the NFIP on a path towards solvency, and provide clearer disincentives for development in the floodplain, they also present serious challenges to the affected people and businesses, who had chosen to live or work in the floodplain under a very different regulatory environment.

New Flood Insurance Rate Maps

Apart from these structural reforms, FEMA has also prioritized the updating of FIRMs, some of which are over 40 years old. Due to a combination of sea level rise, ground subsidence, and expansion of impervious surfaces, analysts expect significant net expansions of the SFHA. This expansion of the SFHA is concerning first and foremost because it represents increased risk of harm to people’s lives and property, but also because the insurance requirements within the SFHA can be very burdensome for people already experiencing financial or social problems.

Population Estimation

In order to target assistance for those who are negatively impacted by the expansion of the SFHA, it is necessary to first understand the characteristics of the affected populations and properties. This requires the integration of demographic data, typically collected by the Census or other survey, with the flood zones, which of course cut across census boundaries. The Key Readings for this week focus on problems related to the estimation of population characteristics, and underscore how important the chosen methodology is.

Key Readings:

1. Fielding, J., & Burningham, K. (2005). Environmental inequality and flood hazard. Local Environment, 10(4), 379–395. doi:10.1080/13549830500160875. (called Fielding_Environmental_inequality_2005.pdf on website and Stellar)
2. Maantay, J., & Maroko, A. (2009). Mapping urban risk: Flood hazards, race, & environmental justice in New York. Applied Geography, 29(1), 111–124. doi:10.1016/j.apgeog.2008.08.002. (called Maantay_Mapping_urban_risk_2009.pdf on website and Stellar)

Additional References:

1. Findlan, K., Vaghela, V., Weselcouch, M., & Yager, J. (2014). The Price of Resilience: Can Multifamily Housing Afford to Adapt? NYU Furman Center. http://furmancenter.org/thestoop/entry/the-price-of-resilience-can-multi-family-housing-afford-to-adapt
2. FEMA. (2011). Answers to Questions About the NFIP. http://www.fema.gov/media-library/assets/documents/272
• Note: this is a good primer on the NFIP, but does not reflect the 2012-2014 reforms.
3. FEMA. (2014). How Recent Legislative Changes Affect Flood Insurance. http://www.fema.gov/media-library/assets/documents/95946

Discussion questions:

1. How did Fielding & Burningham (2005) justify their claim that the grid method of population estimation was more valid than the census district method? Given the contradictory results, are there other studies you have read that you would question?
2. How does Maantay & Maroko (2009) improve on population estimation methods?
3. Are there are other examples you can think of for using fine-grained data to increase the resolution of coarse-grained data?
4. What population or property characteristics would be most important to consider when looking at those impacted by the expansion of the SFHA? What methods would you suggest for the analysis?

Back to 11.522 home page.