Climate Change and Homeowner's Insurance In Florida

# Climate Change and Homeowner's Insurance In Florida  ## Introduction, Through Article Excerpts Solvency issues are material (also from [here](https://www.fox13news.com/news/climate-change-rocking-the-insurance-industry-and-homeowners-feel-the-heat)): > In May 2021, the state allowed three companies to drop home insurance policies for more than 50,000 homeowners in an effort to help the companies to stay afloat financially — facing millions in reported net losses. Insurance experts told FOX 13 Tampa that increasing hurricane claims, roofing repair schemes and lawsuits were all to blame. > But nonetheless, several of the state’s smaller insurers have recently gone out of business, leaving thousands of Floridians with no option other than Citizens, the government-run insurer of last resort. There is precedent for consolidation and market exit resulting from natural disasters e.g. the events leading to the formation of the insurer of last resort in Florida (see [here](https://en.wikipedia.org/wiki/Citizens_Property_Insurance_Corporation)): > Hurricane Andrew in 1992 was the costliest storm the United States had experienced, with $26.5 billion in damage. It took a huge bite out of the reserves for claims held by 30 insurance companies doing business in Florida. Eleven insurance companies were bankrupted, while others stopped writing or renewing property insurance policies in the state. Those that remained raised premiums and deductibles across the board and limited the number of high-risk policies they wrote. Almost 1 million coastal homeowners were unable to find any company willing to insure their homes, so the Florida Legislature authorized the formation of the Florida Residential Property and Casualty Joint Underwriting Association (FRPCJUA) and the Florida Windstorm Underwriting Association (FWUA) as the insurers of last resort. There are constraints on how the insurer of last resort can "compete" with private insurers (see [here](https://en.wikipedia.org/wiki/Citizens_Property_Insurance_Corporation)): > Through 2006, Citizens Insurance charged its customers the highest rate approved by the Florida Office of Insurance Regulation to avoid competing with private carriers. Insurance agents were prohibited from writing policies through Citizens if there was a private (not surplus lines) carrier that would write the risk. If a qualified insurance company was willing to take a group of policies, Citizens Insurance would transfer them to that company and cancel coverage. Customers had no recourse. > Florida Senate Bill 2498, known as the Glitch Bill, was signed into law by Governor Crist on June 11, 2007. This legislation permitted agents to write a Citizens policy for customers if the premium for a comparable policy offered by a private carrier was 15% (instead of 25%) more expensive. Customers were also allowed to stay with Citizens Insurance if they were notified that their policy was being assigned to a private carrier. There are different business models at play (see [here](https://en.wikipedia.org/wiki/Citizens_Property_Insurance_Corporation)): > Since the 2005 pull back by insurance giants Allstate, State Farm, etc., small, in-state companies have been taking a larger share of policies. These start-ups have not followed the traditional insurance model by accumulating cash reserves to cover expenses in high claim years. Instead, they pay as much as half of the policy premium for reinsurance to offshore companies to cover claims. In the seven years since the last major Florida hurricane, profits have risen, but many small companies shifted that money into affiliated businesses and ignored the need for a reserve. Lots of recent events are relevant: > Following Hurricane Ian in October, Citizen's policy count exceeded 1.1 million and the company was projected to insure 15% of the market by end of 2022. From early 2022 to the first quarter of 2023, seven Florida insurers had been declared insolvent including United Property & Casualty Insurance (UPC) with 135,000 policies. Their policies were concentrated in Southwest Florida and experienced 25,000 claims with losses of $864 million from Hurricane Ian. The Florida Insurance Guaranty Association board met on March 31 and filed for an emergency assessment of 1% on all Florida property insurance policies. That is in addition to a 0.7% for 2022, a 1.3% assessment from July 1 2022 to June 30 2023, and another 0.7% ending December 31 2023. As of April 7, 2023 Citizen's policies numbered 1,248,000. Citizens Insurance Board of Governors submitted a 14.2% rate increase on March 31, 2023 effective in November. However, Florida statutes limit rate increases on homesteaded property to 12% per year. A comment on private insurer exits socializing risk and premium increases being constrained, from [this report](https://fc100.org/into-the-storm-framing-floridas-looming-property-insurance-crisis/) on Florida's Insurance Crisis: > "[...] leaving the state vulnerable in the interim while continuing to provide deeply discounted insurance to millions of property owners and shifting tens of billions of dollars of risk from a market-based private sector to taxpayers." ## Potential Questions 1. There's a wedge between the rate at which premia should rise, accounting for increased climate risks, and the rate at which premia can rise due to regulatory frictions (from Katherine Wagner; see [here](https://krhwagner.com/papers/ClimateChangeInsurance.pdf)). See below. One potential question might be: what would be the market structure implications of allowing premia to rise in line with risk?  > "Previous attempts at mandating flood insurance rate increases in this market were so unpopular that they were ultimately rescinded. Insurers from California to Germany are reducing coverage for homeowners they perceive as uninsurable at current prices. The slow updating of contract pricing is partially due to statutory price caps in some markets: Californian wildfire premium increases in excess of 6.9% trigger extensive regulatory reviews. Many insurance models underestimated the rapid escalation of climate change impacts, and the lag in updating premiums to reflect new information has created solvency difficulties for some insurers as costs have soared." > "In addition to the mismatch between the rates of change of risk and price, premium levels also typically weren’t set with climate change in-mind. Many natural disaster insurance markets set rates using backward-looking formulas that exclude future changes in risk. Private wildfire insurers in California must set rates based on their own losses over the past twenty years; using model-based predictions of future risk is actually illegal." 2. Should there be a trade-off between actuarial fairness and enabling natural disaster insurers to build up reserves? See below (from Katherine Wagner; see [here](https://krhwagner.com/papers/ClimateChangeInsurance.pdf)). > "Combined, these features suggest that regulation of natural disaster insurance markets may need to be approached differently than other insurance markets. For example, policies that mimic other markets by implementing actuarially fair premiums may not achieve their objectives here since such premiums may not allow insurers to accumulate reserves for catastrophic loss years or may not not be viewed as worthwhile by homeowners who misperceive their risk." 3. One industry report suggests incentivizing mitigation measures through insurance premium discounts, as "for every dollar spent in storm preparation, $4 to $7 is saved in reconstruction costs" (see [here](https://fc100.org/into-the-storm-framing-floridas-looming-property-insurance-crisis/)). Could we look into the interplay in incentives / the amount of savings passthrough? 4. At one point the insurer of last resort was offering to pay private insurers to take on their customers (see [here](https://en.wikipedia.org/wiki/Citizens_Property_Insurance_Corporation) and below). (Indeed, this is not just an historical artifact but [Florida statute](http://www.leg.state.fl.us/statutes/index.cfm?App_mode=Display_Statute&URL=0600-0699/0627/Sections/0627.3511.html) as of 2022.) Could one study whether this is welfare-enhancing / what the market structure impacts were? What are the benefits from risk-sharing between the public and private sectors? > In May 2013, just three of the eight members of the Citizens Board of Governors approved a deal in which Heritage Property and Casualty Insurance, formed in 2012, was offered $52 million to take over 60,000 accounts ($867 per policy). Florida CFO Jeff Atwater admitted that Heritage did not have the financial strength to take over the policies without the incentive, but Citizen's President Barry Gilway insisted that Heritage was "one of the most well-capitalized" in Florida. > New insurance companies in Florida must be approved by the Office of Insurance Regulation (OIR), which reviews the firm's business plan, projections and financial condition. The judgment of the OIR has been questioned because six of the 18 companies licensed to write "takeout" policies between 2007 and 2011 failed, even though Florida had no major hurricanes during that time frame. Florida taxpayers were forced to cover $400 million in losses. Despite concerns about the OIR, "State regulators have approved six property insurers to remove up to 151,000 policies from state-run Citizens Property Insurance Corp. in February [2014]. [...] Florida's Office of Insurance Regulation last month approved First Community Insurance Co. to take out as many as 51,249 Citizens policies, while Safepoint Insurance Co. may remove up to 40,000 policies. Elements Property Insurance Co. and Heritage Property Casualty Insurance Co. each have been approved for up to 20,000 policies. Southern Fidelity Insurance Co. and Southern Fidelity Property & Casualty can remove 10,000 each. Private insurers have taken out more than 312,000 Citizens policies this year. [Citizens] President Barry Gilway has said the goal is to trim Citizens down to about 800,000 policies." 5. Katherine Wagner notes that the US is fairly unique in that e.g. flood insurance is sold separately from insurance for other perils (see [here](https://www.aeaweb.org/articles?id=10.1257/pol.20200378)). Why is this, and what would be the risk sharing implications of bundling insurance for different kinds of natural disaster? ## Related Work A few starting points are: 1. Adaptation and Adverse Selection in Markets for Natural Disaster Insurance (Katherine Wagner): Tests and finds no evidence for adverse selection in flood insurance take-up. Estimates willingness to pay for flood insurance and finds low WTP; attributes the wedge between expected payouts and WTP to behavioral factors, in particular underestimates of flood risk. Finds that increasing premia to actuarially fair prices without insurance mandates would be welfare-decreasing due to low WTP (and resulting lower levels of insurance). 2. Pricing of Climate Risk Insurance: Regulation and Cross-Subsidies (Oh et al.): Constructs a measure of price setting frictions for insurers by state, using regulatory filings that record prices requested by insurers and realized (regulator-permitted) prices. Shows that price setting frictions are positively correlated with climate risk and that insurers operating in multiple states subsidize higher climate risk, high price-setting friction states through premia in low price-setting friction states. ## Available Data The Florida Office of Insurance Regulation maintains a number of useful-seeming data sources, for example: * [Data](https://floir.com/property-casualty/take-out-companies) on companies purchasing / being sold insurer-of-last-resort policies. * [Data](https://floir.com/resources-and-reports/new-entities-to-the-market) on new entrants to the insurance market. Separately, it might be possible to partially access some of the data used by Katherine Wagner: 1. A dataset on home locations from Zillow, linked to flood risk maps to determine flood risks for each home. 2. The universe of flood insurance policies and claims between 2001 and 2017 (obtained via FOIA requests) in 20 coastal states. Oh et. al's data could potentially also be sourced (at least to some extent): 1. Data on homeowner's insurance premium change filings from the National Association of Insurance Commissioners. 2. Data on climate losses from the Spatial Hazard Events and Losses Database for the United States. 3. Data on the insurers of last resort in each of the (almost all 50) states examined. ## Appendix  Useful links: * A good [overview](https://www.bankrate.com/insurance/homeowners-insurance/florida-homeowners-insurance-crisis/#why) of the situation in Florida. * An [overview](https://www.progressive.com/answers/what-does-homeowners-insurance-cover/) of what homeowner's insurance typically covers.