N Norex Engineering, Inc. O 1220 E. Main Street R League City, TX 77573-4157 E 281-474-2640 Office X 281-474-2748 Fax Harris County September 30, 2013 Public Infrastructure Department Attn: John Blount Director, Architecture & Engineering 1001 Preston, 7 th Floor Houston, Texas 77002 713-755-6888 Re: Biggert-Waters Act of 2012 The following economic impact study analyzing the effect of the Biggert Waters Flood Insurance Reform Act of 2012 (BW 12) on the constituents of Harris County was completed on the behalf of Harris County. Herein is presented a summary of our findings and the economic impact study as prepared by Quanticon, LLC. While there are relatively few constituents immediately affected (approximately 0.2% of all residences) by BW 12, those affected may see dramatic increases to their flood insurance premiums. Effective January 1, 2013, non primary residential structures in a Special Flood Hazard Areas (SFHA) receiving subsidized Pre Flood Insurance Rate Map (FIRM) rates will experience an increase in their annual premiums of 25% until the full risk rate is reached. In approximately 4 years, these homeowners will see their premiums effectively doubled from 2012 rates. Non primary residences represent roughly 15% of all residences in higher risk flood zones receiving subsidized rates. Effective October 1, 2013, the elimination of subsidies for severe repetitive loss properties consisting of 1 4 residences, business properties and properties that have incurred flood related damages where claims payments exceed the fair market value of the property will begin. New and lapsed policies will be issued at full risk rate. FEMA continues to implement BW 12, and further updates providing detailed information on Preferred Risk Policy Extensions and Grandfathering flood insurance policies are anticipated. Please find attached the Economic Impact Study, Harris County Precinct 2 Coastal Regions 2013, dated September 2013, as prepared by Quanticon, LLC. The following is a summary of the study s findings regarding the effect of changes in flood insurance premiums on economic activity in the study area, a sub-region of Harris County Precinct 2. Loss of real estate value - $270,269,940 Loss of annual Ad Valorem Tax - $5,829,712 Includes $3,264,639 in school taxes Reduction in annual business volume - $43,127,257 Immediate loss - $12,249,147 Loss of Jobs 255 jobs Immediate Loss 75 jobs Annual reduction in personal income - $10,514,422
While the findings above can generally be summarized as outlined, please note that the exposure to localized reductions in property values is particularly severe for Morgan s Point, Nassau Bay, Seabrook, Shoreacres and Taylor Lake Village. Therefore, it can be assumed that the vast majority of the economic impact will be borne by the property owners in these municipalities. Additionally, the study does not reflect all increases in insurance premiums, and select properties will experience substantially higher premium increases than those listed on the example table contained within the study. Economic impacts of the legislation to neighboring geographic regions have not been considered within the scope of this study. Based upon the lack of any provision in the law that allows FEMA to consider affordability as it implements the premium increases required by BW-12, we recommend that it be delayed and amended to include such considerations. We also recommend that grandfathering be reinstated for all properties that were built to the required elevation at the time of construction, have maintained their flood insurance policies, and have not suffered repetitive loss. It is our opinion that permitting jurisdictions adopt new FIRM maps as-is; immediately, regardless if they are modified in the future, to ensure that all new construction will meet or exceed expected base flood elevation. Adoption of the proposed FIRM will result in an increase cost of new construction without the risk of future property value depreciation. Furthermore, we recommend that Harris County undertake a feasibility study aimed at communitywide efforts to develop and construct a comprehensive levee system and other flood control projects within the boundaries of Harris County, inclusive of the Galveston Bay Area. All efforts should be made to apply for and secure a significant portion of the $400 million/year 5-year FEMA funds that under the adoption of BW-12 will be made available to effected communities and private property owners. Please contact my office if you have questions or need any additional information. Sincerely, Michael T. Scanlon, P.E. President Norex Engineering, Inc. MTS/ls
The Effect of Changes in Flood Insurance Premiums on Economic Activity in a Sub-Region of Harris County Precinct 2 An Independent Report Prepared for Norex Engineering, Inc. by Stephen J. Cotten, Ph.D. Jason E. Murasko, Ph.D. of Quanticon LLC September 2013 Copyright 2013 Quanticon LLC. May not reproduced or distributed without the prior written permission of Quanticon LLC.
EXECUTIVE SUMMARY Objective: To evaluate the economic impact to the study region (defined as a sub-region of Harris County Precinct Two) from income and wealth effects related to the effective and proposed removal of flood insurance subsidies and the updating of flood maps related to the Biggert-Waters Reform Act of 2012 (BW-12). The economic impact will be defined by changes to business volume, jobs, personal income, and government revenues. Findings: Our estimated aggregate economic impact to the study region as a result of fully realized premiums increased due to BW-12 is a $43,127,257 reduction in annual business volume and an associated $5,829,712 annual reduction in property tax revenues spread over local governments. Immediate impact is estimated at a $12,249,147 reduction in annual business volume which will increase over time as primary residential premiums increase with property turnover. Exposure to localized reductions in property values is particularly severe for Morgan s Point, Nassau Bay, Seabrook, Shoreacres, and Taylor Lake Village. 1
BACKGROUND The National Flood Insurance Program (NFIP), established by the National Flood Insurance Act of 1968, allows property owners to purchase flood insurance in exchange for community floodplain management designed to reduce future flood damages. Major losses in the last decade, particularly from Hurricanes Katrina and Sandy, have forced the program to borrow $28 billion from the US Treasury. In response, Congress passed the Biggert-Waters Flood Insurance Reform Act of 2012 (BW-12) to restore the solvency of the NFIP by increasing the premium intake of the program, largely by eliminating subsidies. Over 80% of insurance policies in the NFIP have rates that exceed the actuary rate to subsidize the remaining 20% of policies. These subsidies come from two forms of grandfathering. First, structures built before the first effective Flood Insurance Rate Map (FIRM) for their area may receive insurance at pre-firm rates. Second, structures that have been placed into new flood zones as updated maps are developed have the option to continue receiving insurance from the previous risk-rating. FEMA places properties into three general risk zones: X (includes areas of low flood hazard defined as less than a 0.2% chance of annual flooding and areas of moderate flood hazard defined as between a 0.2% and 1% annual chance of flooding; also used for some areas protected by levees from a 1% chance of annual flooding, shallow flooding areas with average depths of less than one foot, or drainage areas less than 1 square mile), A (high-risk areas with a 1% chance of flooding per year from rising water), and V (high-risk coastal areas with a 1% chance of flooding per year from storm waves). Flood insurance purchases are optional for properties in X zones and mandatory for A and V zones for homes and buildings with mortgages from federally regulated or insured lenders. Premiums for X zones are above actuary rates and are based on the amount of coverage purchased. Post-FIRM 2
premiums for A and V zones are also set above actuary rates and require a comparison of a property s Base Flood Elevation (BFE, or the elevation at which flooding is expected at a 1% per year chance) and the surveyed elevation of the lowest floor of the property in A zones or the elevation of lowest horizontal element in V zones. A property built four feet above BFE will generally see premiums not much higher than those for X zones, while properties built under BFE have annual premiums that are a minimum of 1% of the insured value. Pre-FIRM subsidized premiums average approximately $1,200. The magnitude of subsidy is not known with certainty, but estimates place the average subsidy between $1,400 and $1,600 with an implied full-risk premium of $2,600 to $2,800. 1, 2 BW-12 closes the operating deficit of the program by removing subsidized rates for residential properties that are not the primary residence of an individual, properties with a history of flood damage, business property, property that has been substantially rebuilt or improved after passage of the bill, any lapsed or new policy, or any policy for which the owner has refused a FEMA offer to assist in rebuilding the structure to avoid future flooding. Rates may increase by 25% per year until actuarial rates are achieved. BW-12 also allows rates to increase by 20% per year for properties that have higher rates as a result of new maps. BW-12 allocates $400 million per year for the next five years to updating flood maps, which for some regions will move a substantial number or properties from low-risk to high-risk flood zones, bringing new BFE requirements to existing structures that did not have such requirements previously. The structure 1 U.S. Government Accountability Office, Flood Insurance: More Information Needed on Subsidized Properties. GAO-13-607. Available at http://www.gao.gov/products/gao-13-607. 2 Keybridge Research (as prepared for FEMA), Reforming the National Flood Insurance Program: An Evaluation of Policy Alternatives. Available at http://www.fema.gov/pdf/business/nfip/reforming_the_nfip_an_evaluation_of_policy_alternatives.pdf. 3
and timeline for BW-12 implementation can be found at http://www.fema.gov/flood-insurancereform-act-2012. Higher flood insurance premiums affect economic activity in two primary ways. First, increased flood premiums reduce disposable income for consumers and profits for businesses that would otherwise be spent on goods and services in the local economy (an income-effect ). Second, higher insurance premiums reduce property values. This results in a further reduction in spending (a wealth-effect ) and reduces property tax collections for local governments. These two channels affect a relatively small proportion of Harris County properties. In Harris County, there are approximately 270,000 flood insurance policies in effect. 3 For perspective, the county has approximately 1.6 million housing units (of which 64% are single-unit) and nearly 450,000 business establishments. As a percentage the effective flood policies represent about 13% of housing and business units. Of the 270,000 flood policies, only about 8% are subsidized (1% of total housing and business units). 21% of subsidized policies (0.2% of total housing and business units) are business and non-primary residences which are the types of properties facing immediate increases of subsidies by 25% per year to reflect the true risk rate. Despite the small proportion of Harris County properties affected by BW-12 implementation, there are certain localities which because of their proximity to waterways and/or low elevations will experience a disproportionate impact from changes to the NFIP. 3 FEMA: NFIP Policyholders: Total Number of Subsidized Policies by State and County http://www.arcgis.com/home/webmap/viewer.html?webmap=e0208985e8e64d44bca999325254ff5b 4
STUDY OBJECTIVE The study objective is to evaluate the economic impact to the study region from incomeand wealth-effects related to the effective and proposed removal of flood insurance subsidies and the updating of flood maps. The economic impact will be defined by changes to business volume, jobs, personal income, and government revenues. The study region (outlined in dark blue) was defined by Harris County Public Infrastructure Department (HCPID) as a portion of Harris County, Precinct 2: The study region is bounded on the north by I-10E, to the west by the East Loop Freeway and Gulf Freeway, to the east by Cedar Bayou, and to the south by Galveston Bay and Clear Lake. The area encompasses approximately 700 km 2 relative to 2,600 km 2 of all of Harris County and 5
includes localities that were in the highest-risk categories prior to remapping and where remapping will greatly expand high-risk zones. All references to cities and other local districts in this report refer only to their land area within the study region. The primary source of data for this report is a property database provided by HCPID. This database provides information on 159,391 properties in the study region. Data is provided on several characteristics including location by city/address, land use, appraised/market value, and flood zone under the previous and current maps. For some properties there is also information on the year of construction and comparison of surveyed/estimated first-floor elevation (FFE) and BFE. All data were sourced, compiled, and verified by HCPID. 6
SCALE OF REMAPPING For the purposes of this study we define a high-risk zone as any of the zones falling inside the 1% floodplain: AO, A, AE, VE, and AE/VE floodways. Table 1 summarizes the effect of remapping on the counts of properties in high-risk zones, with added categories for singlefamily homes (state code A1) and commercial properties (state code F1): Table 1: Properties in High-Risk Zones Counts Type Total Previous Map Current Map Change All properties 159,391 26,748 31,414 4,666 Single-family homes 113,356 15,948 19,611 3,863 Commercial 8,809 1,644 1,772 128 Percentage of total All properties - 16.8% 19.7% 2.9% Single-family homes - 14.1% 17.3% 3.2% Commercial - 18.7% 20.1% 1.5% Note: All properties includes single-family homes, commercial properties, and other property types such as multi-family residences, industrial property, and utilities. Remapping results in a net positive addition to properties located in high-risk zones. About 2.9% of all properties are added to high-risk zones, including 3.4% of single-family homes and 1.5% of commercial properties. Using exemptions as a proxy for primary/non-primary residences, 81.8% of the change in single-family homes are primary residences (defined by homestead exemption) and 18.2% are non-primary residences (defined by the absence of exemption). 7
IMPACT ANALYSIS Estimate of Potential Premium Increases We evaluate two components of premium increases: 1) new elevation requirements resulting from remapping into high-risk flood zones and 2) the removal of pre-firm subsidies. Previously low-risk properties that remain in low-risk zones are not affected by these sources of premium increases and are therefore omitted from the analysis. Of the remaining properties, we evaluate those that are classified as single-family residences or commercial property, and have positive building values. There are 21,629 properties in the HCPID database that satisfy these conditions. Of these, there are 1,794 commercial properties (8.3% of total), 15,489 single-family primary residences (71.6%), and 4,346 single-family non-primary residences (20.1%). The following table compares flood zone placements for each property type under the previous and new maps: Table 2: Counts of Properties by Type, Previous Flood Zone, and Current Flood Zone Previous Flood Zone Primary Residences Non-Primary Residences Commercial Properties Current Flood Zone Current Flood Zone Current Flood Zone V A 0.2% X V A 0.2% X V A 0.2% X V 267 16 0 0 109 17 0 0 48 15 0 0 A 409 11,480 150 6 161 3,261 62 8 73 1,460 34 2 0.2% 2 3100 0 0 2 694 0 0 0 150 0 0 X 0 59 0 0 0 32 0 0 0 12 0 0 Each cell gives the count of properties in a current flood zone that were in a given flood zone under previous maps. Numbers in the shaded cells represent properties that were not remapped. Cells below the shaded area represent properties that have been remapped into higher-risk zones, while cells above represent properties that have been favorably remapped into lower-risk zones. 8
In total, 262 properties are remapped from high-risk to low-risk, 4,051 are remapped from lowrisk to high-risk, and 17,316 are high-risk in both maps. The tables also indicate a net shift of 595 properties from A-zones to V-zones. Two key variables in the determination of premiums are construction date (to identify pre-firm and post-firm rates) and lowest floor elevation (to determine its relation to BFE). The HCPID database does not provide complete data on either of these variables. Date of construction is provided for 8,331 (38.5%) of our 21,629 properties and complete elevation data is provided for 13,822 (63.9%). We therefore employ the following procedure to estimate aggregate premium changes over all 21,629 properties by extrapolating from the construction date and elevation data supplied by HCPID: 1. The average building value is calculated for each cell in Table 3. For example, the upper-left cell includes 267 primary single-family residences that were placed in a V-zone in both previous and current maps. The average building value for these properties is $103,752. 2. For each average building value, both pre-firm and post-firm premiums are calculated from the current FEMA ratings manual. 4 Post-FIRM premiums for high-risk (A and V) zones are weighted according to the elevation distribution of the HCPID properties that have elevation data. This effectively applies the elevation distribution of properties with elevation data to the entire sample of 21,629 properties. Details on this are provided in an appendix to this report. 4 Available at http://www.fema.gov/flood-insurance-manual. We assume that 100% of properties pay premiums that insure 100% of building values, a 0.4 contents-to-building ratio for residences, and a 1.0 contents-to-building ratio for commercial property. These ratios are based on the relative maximum NFIP coverage for building and contents for each property type. 9
3. We perform a weighted summation of pre-firm and post-firm premiums over the counts of each cell in Table 3. Pre-FIRM properties are identified as those constructed before 1974. The distribution of construction dates for the 8,331 properties with available data has 52% of properties built before 1974 and this percentage is applied to all counts in Table 3. For example, of the 267 primary single-family residences in the upper-left cell, we assume that 139 are pre-firm and 128 are post-firm. We then multiply the appropriate pre-firm average premium by 139 and post-firm premium by 128, and sum the total. 4. Finally, the weighted sum of premiums is scaled to reflect the percentage of property owners who purchase flood insurance. It has been estimated previously that 57.8% of homes in highrisk flood zones have an outstanding federally-related mortgage and are subject to mandatory insurance purchase requirements and that 42.8% of residential properties with purchase requirements actually comply. 2 Therefore, 24.4% of properties are required to buy flood insurance and do so. Adding purchasers of flood insurance who do not have a mortgagemandate, estimates find the penetration rate of flood insurance is 26% for properties in highrisk flood zones and 3% for low-risk zones. 2 We therefore multiply the weighted sum of high-risk premiums by.26 and the sum of low-risk premiums by.03 to arrive at a value for aggregate premiums paid. This aggregating procedure is performed twice: for previous maps with pre-firm subsidies in-force and for current maps under the assumption that all properties pay post-firm rates (i.e. the removal of pre-firm subsidies). The difference between cases yields the reduction in 10
disposable income used for the economic impact analysis. The following table summarizes the results of the premium calculations: Table 3: Estimated Aggregate Premiums for 21,629 HCPID Properties Property Type Previous Current Difference Primary Residential $9,705,707 $18,976,742 $9,271,035 Non-Primary Residential $2,802,977 $5,069,137 $2,266,005 Commercial $7,099,732 $9,076,189 $1,976,457 Total $19,608,416 $33,122,068 $13,513,497 The shaded amount of $13.5 million represents our estimate of an aggregate premium increase owing to remapping and subsidy removal. Estimating the Economic Impact We estimate the economic impact that derives from an income-effect associated with an aggregate increase in flood insurance premiums in the study region and a residential wealtheffect associated with an assumed reduction in property values associated with capitalized insurance premiums. We assume a capitalization rate of 5% resulting in an overall reduction in residential property values of $231 million, of which 2.5% is assumed to translate to a reduction in consumer spending. 5 We employ an economic impact model that incorporates the direct effect from a reduction in local spending and a multiplier effect that arises from continued reductions in spending that spread through the various industries in the study region. 5 We take this as an average wealth-effect associated with housing values. Recent discussions on postrecession housing wealth suggest an effect of around 1%, but previous estimates put this percentage at 3-5%. See Soss, N. & Mo, H. 2013. "U.S. Economics Digest: Honey, I Shrunk the Wealth Effect," Credit Suisse Economics Research, February 13 th, and Sufi, A. 2013. Will Housing Save the U.S. Economy? 2013 Chicago Booth Economic Outlook Discussion Paper. 11
Our estimated economic impact associated with these effects is a $43,127,257 reduction in annual business volume as a result of a fully-realized $13,513,652 aggregate increase in flood insurance premiums. The impact on jobs is estimated at 255 jobs lost with an annual reduction in personal income of $10,514,422.With respect to the impact on local governments, the reduction in both residential and commercial property values due to the capitalization of insurance premiums is $270,269,940. Accounting for exemptions, we estimate this to associate with a $5,829,712 annual reduction in property tax revenues spread over local taxing authorities in the study region, inclusive of $3,264,639 in school district taxes. These estimated economic impacts assume a complete and full realization of the estimated aggregate increase in premiums. It is important to note that the whole of the aggregate premium increase will not be realized immediately. The timeline of BW-12 has initial premium changes occurring for non-primary residences and commercial properties. Primary residences will incur premium shifts upon sale of the property, lapse in existing policy, repeated/severe flooding event, refusal to mitigate for flood risk, or purchase of a new policy. The initial premium changes associated with non-primary residences and commercial property is associated with a $12,249,147 reduction in annual business volume in our model, 73 jobs lost, and a $2,986,341 reduction in annual personal income. 6 The full economic impact will be realized over time at a rate determined by turnover of primary residential homes. Assuming a 6% annual turnover that is unaffected by changes in insurance premiums, then one-half of the full impact will be realized after 5 years and two-thirds of the impact will be realized after 12 years (Figure 1). 6 Though premium increases for a given property are phased-in over 5 years, we assume that property owners and purchasers behave according to the expected full increase in premiums. 12
Figure 1: Time to Realized Impact 0.9 0.8 0.7 % of Impact 0.6 0.5 0.4 0.3 0.2 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Years A number of factors may reduce the economic impact. These include: A 6% annual turnover in property is an average. Some property owners will stay in their homes for decades. These non-transferred properties will not experience premium increases per the existing language of BW-12. Not all pre-1974 properties are pre-firm. Properties that have undergone substantial remodeling or reconstruction are required to pay post-firm rates. Furthermore, property owners may opt to obtain an elevation certificate and pay post-firm rates if they believe their elevation sufficiently high to reduce premiums. The elevation distribution of properties with known values may not be representative of all 21,629 properties. In a statistical comparison of properties with and without HCPID elevation data, there were statistically significant (probability values less than 1%) higher propensities for properties with data to fall into V-zones relative to A-zones. This ranged 13
from a 1.5% difference for residential properties to a 6.5% difference for commercial properties. 7 Estimated premiums for properties well below BFE may be overstated. We are forced to predict certain premium rates due to the absence of published NFIP rates for properties below a certain BFE threshold (2 feet under for A properties, 4 feet under for V properties). This procedure and its potential for overstated rates are described in the appendix. One factor that has an ambiguous effect on our estimates is that local insurance purchase rates may not reflect the national average. As stated previously, national estimates show that approximately 26% of properties in high-risk flood zones have flood insurance. 42.2% of properties do not have mortgages and are not required to carry insurance. 57.2% of property owners facing mandatory purchase requirements fail to comply. If individuals in Harris County are more (less) likely to buy optional insurance and/or are more (less) likely to comply with flood insurance requirements, this would increase (decrease) the impact of increased insurance premiums. Furthermore, we hold the compliance rate constant; as premiums increase it will likely exert downward pressure on purchase rates. This impact analysis is an estimate of the aggregate impact on BW-12 implementation on the whole of the study region. However changes to the NFIP program may have disproportionate effects on localized property values and influence the consequent behavior of property owners and purchasers in those localities. This issue is addressed in the following section. 7 We are less concerned with the distribution of construction dates. Commercial properties exhibited no statistical differences between those with and without data. Residential properties with data exhibited significantly higher propensities to V-zones relative to those without data, but at less than 1%. 14
IMPACT ON LOCALITIES Home elevation and BFE changes are highly spatially correlated, leading to clustering of both positive and negative effects. Table 4 provides information by city in the study area. As indicated by the table, the cities of El Lago, Nassau Bay, and Taylor Lake Village are all substantially affected by remapping, with 30-40% of properties being remapped to high-risk flood zones. Seabrook also has about 14% of properties remapped to high-risk. After remapping, the cities of El Lago, Nassau Bay, Seabrook, Shoreacres, and Taylor Lake Village all have 70% or more of their properties located in high-risk zones, generally reflecting single-family residences. These localities may be indicated as potentially most sensitive to changes associated with BW-12. Table 4: Properties in High-Risk Zones, By City Counts of Properties Previous Map Current Map Change City All A1 F1 All A1 F1 All A1 F1 Baytown 3,582 1,890 151 3,779 2,092 147 197 202-4 Channelview 354 67 54 389 70 63 35 3 9 Deer Park 708 475 78 718 475 78 10 0 0 El Lago 606 271 15 1,090 726 17 484 455 2 Galena Park 72 29 3 204 141 6 132 112 3 Houston 4,595 2,968 215 4,724 3,046 239 129 78 24 Jacinto City 27 13 4 29 14 5 2 1 1 La Porte 2,997 1,756 160 3,504 2,053 174 507 297 14 Morgan's Point 85 51 0 89 51 2 4 0 2 Nassau Bay 974 800 19 1,455 1,172 44 481 372 25 Pasadena 5,204 3,454 324 6,466 4,426 359 1,262 972 35 Seabrook 3,814 2,308 183 4,537 2,847 195 723 539 12 Shoreacres 801 572 2 801 572 2 0 0 0 South Houston 1,750 877 414 1,783 905 414 33 28 0 Taylor Lake Village 447 382 3 1,054 958 4 607 576 1 Webster 678 32 18 732 60 22 54 28 4 15
Percentage of Total Properties Previous Map Current Map Change City All A1 F1 All A1 F1 All A1 F1 Baytown 14.3 11.1 10.5 15.1 12.3 10.2 0.8 1.2-0.3 Channelview 21.5 11.0 17.3 23.6 11.5 20.2 2.1 0.5 2.9 Deer Park 6.5 5.0 18.0 6.6 5.0 18.0 0.1 0.0 0.0 El Lago 47.6 31.3 88.2 85.6 83.9 100.0 38.0 52.6 11.8 Galena Park 2.1 1.2 2.4 6.1 5.6 4.8 3.9 4.5 2.4 Houston 10.9 9.7 10.8 11.2 10.0 12.0 0.3 0.3 1.2 Jacinto City 0.9 0.6 2.4 1.0 0.6 3.0 0.1 0.0 0.6 La Porte 20.8 18.3 24.5 24.3 21.4 26.7 3.5 3.1 2.1 Morgan's Point 24.0 41.1 0.0 25.1 41.1 33.3 1.1 0.0 33.3 Nassau Bay 61.2 61.7 35.8 91.5 90.4 83.0 30.2 28.7 47.2 Pasadena 12.8 11.4 14.4 16.0 14.6 16.0 3.1 3.2 1.6 Seabrook 73.4 74.6 89.3 87.3 92.1 95.1 13.9 17.4 5.9 Shoreacres 88.8 100.0 100.0 88.8 100.0 100.0 0.0 0.0 0.0 South Houston 33.4 30.0 55.7 34.0 30.9 55.7 0.6 1.0 0.0 Taylor Lake Village 29.3 27.4 50.0 69.0 68.8 66.7 39.7 41.4 16.7 Webster 31.5 5.6 4.5 34.1 10.4 5.5 2.5 4.9 1.0 Note: A1 = Single-family residences, F1 = Commercial property These percentages in Table 4 are not sufficient to evaluate local impact. Flood insurance premiums in high-risk areas are highly dependent on the relative elevation of the home to BFE. As an example provided by FEMA, a pre-firm single-story residence with $250,000 in building coverage, $1,000 deductible, and $100,000 in contents coverage in an A zone would pay $3,600 per-year without an elevation certificate regardless of property elevation. However, after subsidies are eliminated, this property would pay $553 if four feet above base-flood elevation, $1,815 if at base-flood elevation, and $10,723 if four feet below base-flood elevation. Table 5 shows the distribution of properties in the high-risk flood zone categorized by the difference between FFE and BFE as provided in the HCPID database. Cities such as Pasadena would see modest changes in flood insurance premiums compared to communities such as Seabrook, 16
Shoreacres, and Nassau Bay that have sizable fractions of their properties mapped at more than two feet below BFE. Table 5: Difference from BFE, By Selected City FFE Relative to BFE City < -4-3 to -4-2 to -3-1 to -2-1 to 1 1 to 2 2+ Database Properties % Under -2ft Baytown 153 54 81 75 642 58 19 24,979 1.2% La Porte 299 138 98 124 531 151 34 14,436 3.7% Morgan's Point 5 8 3 6 3 3 0 354 4.5% Nassau Bay 293 65 90 168 297 26 10 1,591 28.2% Pasadena 34 12 59 265 1,825 443 198 40,509 0.3% Seabrook 508 198 240 443 928 69 2 5,195 18.2% Shoreacres 426 84 34 5 0 0 0 902 60.3% Taylor Lake Village 71 24 45 86 336 10 4 1,528 9.2% Note: FFE = first floor elevation (as provided in the HCPID database), BFE = base flood elevation. The percentage of properties at 2 feet or more below BFE is relative to total properties in the database, of which many properties do not have elevation data. Changes in flood insurance premiums can affect the real estate market in several different ways. The expectation is that the increased premiums would be capitalized into property values, causing decreases in value for affected properties with some offset of increased values for surrounding unaffected property. For owners of primary residences with no intention to sell their homes, this is not an issue as subsidized premiums will continue. For property owners who do wish to sell but still retain equity in the property after the value decrease, the increased flood insurance premiums represent a decline in housing wealth and would be reflected in the wealtheffect estimated in our impact analysis from the previous section. However, for property owners with sizable mortgage balances, large flood insurance premiums have the potential to decrease the property value below the remaining value of the mortgage; this would be expected to decrease property turnover as property owners may not be able to afford to sell. 17
One way to augment the impact analysis for the latter case is to increase the assumed percentage associated with a wealth-effect. The idea is that severe property value reductions will have a larger wealth-effect than more modest reductions, for example as underwater owners walk away from their properties. The following table presents estimated economic impacts associated with fully realized estimated increase in aggregate premiums for varying levels of wealth-effect beyond the 2.5% assumed in the main analysis: Table 6: Estimated Aggregate Economic Impact with Varying Levels of Wealth-Effect Percentage Wealth-Effect Impact Outcome 5% 7% 10% 15% Annual business volume $56,851,219 $67,829,992 $84,298,151 $111,745,084 Jobs lost 337 402 499 662 Annual personal income $13,860,322 $16,536,946 $20,551,882 $27,243,442 Note: All values are reductions. The percentage wealth-effect is the reduction in consumer spending expressed as a percentage of overall property value reduction. Wealth-effects beyond 5% would be considered very strong to extreme. This is in part because many properties with large premium increases will benefit from mitigation efforts. For properties with premium increases greater than $5,000 per year, it would be economical to borrow as much as $100,000 to elevate the home (based on a 5% capitalization rate). This expense reduces the wealth of the property owner, but represents income to local construction companies, increasing economic activity within the region and maintaining the property s value and thus minimizing the effect on the property tax base. To the extent that they are made available, FEMA mitigation grants can offset some of these costs of lifting these properties. Quantifying the effect of mitigation on our economic impact is difficult for two reasons. First, the cost to lift a particular structure will vary. Estimates provided by Norex Engineering, Inc. suggest the cost to lift an 18
existing residential structure may be $50,000 to over $150,000. Second, the lifting of a particular structure may not be feasible, for example with a large commercial structure. These costs and conditions would need to be surveyed on an individual-property basis which is beyond the scope of this study. Where mitigation fails and properties are forced into strong reductions in value, impacts to local property bases and consequent tax revenue can be severe. For some perspective on localized exposure, Table 7 gives the total value of properties under BFE for each city and the percentage of that value relative to the value of all database properties for each city: Table 7: Value of Properties Under BFE City Total Market Value % of Total Baytown $52,170,542 1.0% La Porte $76,445,166 1.9% Morgan's Point $15,533,446 20.2% Nassau Bay $193,310,684 36.0% Pasadena $91,663,323 1.0% Seabrook $335,717,249 29.4% Shoreacres $80,608,001 88.2% Taylor Lake Village $97,157,747 28.2% Cities including Morgan s Point, Nassau Bay, Seabrook, Shoreacres, and Taylor Lake Village have 20% or more of their properties exposed to potentially strong reductions in value if mitigation efforts fail. Shoreacres in particular has almost 90% of property value exposed. 19
SUMMARY Our estimated aggregate economic impact to the study region as a result of fully realized premium increased due to BW-12 is a $43,127,257 reduction in annual business volume and an associated $5,829,712 annual reduction in property tax revenues spread over local governments. Immediate impact is estimated at a $12,249,147 reduction in annual business volume which will increase over time as primary residential premiums increase with property turnover. Exposure to localized reductions in property values is particularly severe for Morgan s Point, Nassau Bay, Seabrook, Shoreacres, and Taylor Lake Village. 20
APPENDIX This appendix describes the calculation of post-firm premium rates for A and V zones with BFE requirements. Since the elevation data is not provided for all properties, it is necessary to extrapolate from the available data to all properties evaluated in this report. The following is an example rate table from the FEMA Flood Insurance Manual available at http://www.fema.gov/flood-insurance-manual. We perform similar procedures for A-zone contents coverage and V-zone tables. Source: FEMA Flood Insurance Manual, effective May 1, 2013, Section 5, page 4. The table describes the premium rate structure for building coverage on post-firm properties in A-zones. Rates are expressed as #/#, where the first value refers to insured amounts up to basic coverage ($60,000 for residential properties, $175,000 for commercial) and the second value refers to additional amounts between basic and maximum coverage (up to $250,000 for residential properties, $500,000 for commercial). The table is incomplete in that it does not provide rates for properties whose lower floor elevation is 2 or more feet below BFE (*** refers to Submit for Rating meaning that special 21
calculations are required for these properties). The observed rate structure is nonlinear with respect to decreasing elevation, so we fit a polynomial function to the observed structure as a means to predict the rates for properties at lower elevations relative to BFE. This may overstate rates associated with very low FFE relative to BFE as the fitted function exhibits an increasing rate of change. We make some attempt to correct for this by limiting how far we predict below BFE and assigning these rates to properties below this point. We only consider the first columns for 1 FLOOR properties. For this particular table our predicted rates for 1-4 Family Residences are: Elevation Difference Basic Coverage 4 0.24 0.08 3 0.30 0.08 2 0.42 0.08 1 0.71 0.10 0 1.78 0.13-1 4.40 0.97-2 8.23 2.42-3 11.65 3.36-4 15.66 4.47-5 20.27 5.72 Additional Coverage To arrive at a weighted average that can be applied to average building values in our analysis, we weight each row by the respective percentage of our 21,629 properties (by type) that are estimated to be in that elevation range. For example, the elevation distribution of residential (A1) properties for which HCPID provided data and which are included in our sample is: 22
Elevation Difference Count % of Total 4 71 0.6% 3 52 0.4% 2 165 1.4% 1 1,072 8.8% 0 6,438 53.1% -1 1,543 12.7% -2 718 5.9% -3 627 5.2% -4 638 5.3% -5 810 6.7% Total 12,134 100.0% We apply this to our estimated rate structure as follows: Elevation Difference % Basic Basic % Additional Add'l % 4 0.6% 0.24 0.0014 0.08 0.0005 3 0.4% 0.30 0.0013 0.08 0.0003 2 1.4% 0.42 0.0057 0.08 0.0011 1 8.8% 0.71 0.0627 0.10 0.0088 0 53.1% 1.78 0.9444 0.13 0.0690-1 12.7% 4.40 0.5595 0.97 0.1233-2 5.9% 8.23 0.4868 2.42 0.1430-3 5.2% 11.65 0.6018 3.36 0.1738-4 5.3% 15.66 0.8234 4.47 0.2348-5 6.7% 20.27 1.3528 5.72 0.3819 Weighted average = 3.4871 0.7546 The weighted average is given by the column sums of the adjusted basic and additional coverage rates. This same procedure was applied to A-zone contents rates, V-zone building rates, and V- zone contents rates. Rates were calculated for residential and commercial (non-residential) properties. 23