“Do Battle with Insurance”: Post-Hurricane Insurance Litigation in Southwest Louisiana

  • RSF: The Russell Sage Foundation Journal of the Social Sciences
  • July 2026,
  • 12
  • (4)
  • 178-203;
  • DOI: https://doi.org/10.7758/RSF.2026.12.4.08

Abstract

Property insurance is essential for hurricane recovery, yet timely and fair claims payouts are not guaranteed. Post-disaster litigation over denied or delayed claims is common but understudied in who sues and how this varies sociodemographically. Southwest Louisiana is well situated for such an analysis, where within two years after Hurricane Laura and Hurricane Delta hit the region in 2020, one in five federal civil litigation cases were filed, most regarding insurance. We collate 5,476 court records to geocode implicated properties and assess litigation patterns using property and sociodemographic data. We find litigation is 43 percent more likely in higher percent Black residential areas, with property value amplifying this disparity. Additionally, case duration is longer for residents in these areas, suggesting systemic procedural inequities. Our findings reveal how race, class, and geography relate to insurance litigation, a critical yet overlooked facet of procedural vulnerability that may contribute to inequitable disaster recovery outcomes in the United States.

Property insurance is a critical backstop for financial losses from damaged property from weather- and climate-related hazards in the United States (Kousky 2019). Yet, disasters bring homeowners into contestation with their insurers through litigation—lawsuits homeowners bring against their insurers—to resolve disputed claims. Despite the security property insurance guarantees in the event of a catastrophic loss, insurance companies may undervalue losses, incorrectly adjust or deny claims, or delay payouts, violating insurance contracts and altogether deferring residents’ home reconstruction (Dulin 2008). Delayed housing recovery is consequential for impacted residents’ livelihoods, including housing stability, employment and educational activities, and financial well-being (Fussell 2015; Peacock et al. 2014). Insurers’ “bad faith” practices in delaying payouts for home repairs can be rampant and even expected among consumer groups and insureds (Feinman 2010). Following disasters, resolving property claims disputes through litigation is a common legal need (Finger 2022). Among the array of civil legal issues the American public encounters, insurance contract disputes are a common type of litigation. Yet, many affected individuals may not recognize insurers’ malfeasant actions as lawful or as matters that can be resolved through legal pathways (Sandefur 2014). Understanding these processes is urgent, as minority and low-income populations face barriers or are hesitant to participate in the civil justice legal system in post-disaster contexts (Simington 2023).

Despite well-established evidence of insurance’s role, sparse public administrative records on who is insured impede characterizing the adequacy of insurance for advancing recovery and for whom (Kousky 2019; Squires et al. 2001; You and Kousky 2024). Moreover, a lack of available data prevents documenting any disparate impact or discriminatory practices of insurance companies (Schwarcz 2019; Squires 2003). Given that insurance claims lawsuits are among the most prevalent types of suits on the federal civil court docket following large disasters, these processes have yet to be subjected to social scientific analysis, and the extent of socioeconomic differences in insurance litigation participation is unclear. The role of property insurance litigation in enabling recovery and for whom is a prescient question, particularly given what the pattern foretells amid frequent, severe loss events and narrowing coverage access. The high frequency of insurance litigation following storms raises an important question: Who initiates these lawsuits? This is particularly pertinent as severe weather events become more frequent and insurance coverage narrows.

This gap stands amidst the role of the legal system as a key backdrop to social life and behavior (Friedman 1975), yet one that is undertheorized in how legal participation or lack thereof engenders inequities among impacted residents following disasters. Understanding the policies, planning processes, and institutions that create and sustain racial, class, housing tenure, and gender-based inequalities in recovery following disasters is needed to elucidate why and how inequities occur (Rivera et al. 2022). Procedural vulnerability refers to the disadvantages individuals or groups face due to inequities in processes, systems, and structures (Veland et al. 2013). In this research, we use procedural vulnerability to structure our hypotheses and to motivate insurance litigation as another apparatus through which socioeconomic differences persist post-disaster.

To begin to unpack this dynamic, we present the first study, to our knowledge, to analyze geolocated individual litigation against insurers and examine variation in litigation incidence across property type, demographics, and hazard magnitude. While previous research has mapped and examined the socioeconomic characteristics of individuals in civil litigation cases, such as eviction proceedings (see Hepburn et al. 2020), we do so specifically in insurance suits arising from post-hurricane claims. Our findings have significant policy implications, as states have recently instituted reforms to curb litigation.

There are few places in the United States where insurance litigation is as prominent as Southwest Louisiana, where Hurricanes Laura and Delta struck in the fall of 2020. Within two years, one in five federal civil cases across the US were filed in the United States District Court for the Western District of Louisiana, most of them insurance suits (TRAC 2022). This study interrogates whether and to what degree socioeconomic differences exist in litigation participation, and, among those who participate, the characteristics of the plaintiff, the date of case filing, and case duration. We analyze 5,476 insurance litigation suits tied to disputed claims from Hurricanes Laura and Delta.

In the absence of an empirical summary of these patterns, this study makes three significant contributions to current understandings of property insurance claims participation and the timing of claims following large hurricane events in the United States. First, we present the first empirical, case-level analysis of insurance litigation following a major US disaster, achieved by collating litigation court records. Second, we use statistical regression to evaluate the probability of litigation in Southwest Louisiana, finding that litigation is more likely in higher-percentage Black residential areas and that this effect increases with property value. Importantly, our analysis suggests that the litigant type, as a resident or nonresident, relates to differences in when cases are filed and how long litigation cases last from filing to closure, or settlement.

Our findings suggest that, while residents of diverse socioeconomic, racial, and ethnic backgrounds filed lawsuits against their insurers, plaintiffs are more likely to live in a high-percentage Black residential area, and that income is not a strong predictor of litigation. Instead, property value strengthens this disparity. Who you are as a plaintiff, be it a resident or commercial business, also relates to differences in the length of cases, with residents in higher percent Black areas engaged in cases that last longer. Notably, we are unable to directly measure and account for property-level damage, closed insurance claims that did not result in litigation, or an individual- or group-level indicator of willingness or ability to participate in legal action. So, while our findings suggest that typical plaintiffs in this case are more likely Black or own a higher-valued property, our findings cannot surmise if this pattern is linked to racially motivated practices, such that Black residents experienced discriminatory claims handling or are more inclined to engage in legal actions. These findings advance empirical approaches to document patterns of litigation participation, motivating further hypothesis-driven interrogation of how race, property ownership, and socioeconomics shape legal claims-making in post-disaster contexts.

THEORETICAL BACKGROUND

Situating insurance litigation within broader structures of inequality requires attention to how race and class shape both access to insurance and the capacity to seek legal redress. We draw on multiple bodies of literature on discriminatory insurance practices, the role of law in post-disaster governance, and the social dimensions of civil litigation participation to motivate our theoretical orientation of procedural vulnerability in shaping litigation participation.

Differential Treatment in Insurance Claims Handling

Homeowners insurance is a form of privatized social welfare that protects policyholders against losses from damaging hazard events. Like other forms of social protection, homeowners’ insurance coverage gaps are highest in communities of color and low-income communities (Cornelissen et al. 2024). Leigh Johnson (2015) refers to the resulting protection gap as “splintering protectionism,” in which insurance serves to distinguish the haves from the have-nots in disaster preparedness and recovery (Peacock and Girard 1997). Barriers to accessing insurance, including property insurance, can undermine the ability of homeowners of color to protect their property assets (Squires 2003). Likewise, discriminatory claims-handling practices have been documented for low-income, minority groups (Baker and McElrath 1996; Bolin and Bolton 1986; Lin 2022). Weak and lax insurance regulation can perpetuate discriminatory practices. Federal protection against redress for alleged discriminatory acts is the “Disparate Impact Rule” of the Fair Housing Act, issued by the Department of Housing and Urban Development (HUD) in 2013. The rule defined the term disparate impact as the production or reinforcement of segregated housing patterns. Although the rule can be used to protect against discriminatory insurance practices, it has rarely been used because of the high costs of litigation, and oversight of insurance companies has been weakened and then reinstated under the first Trump and Biden administrations, respectively (Kousky and French 2023). Under the second Trump administration, this rule faces a more tenuous future. Despite the high costs associated with class action lawsuits, the need for redressing discriminatory practices exists, as demonstrated by a recent survey among Black and White policyholders that exposed racial discrimination in claims handling, which has led to an ongoing class action lawsuit against State Farm (Flitter 2022).

Post-Disaster Legal Claims and Insurance Litigation

Insurance companies seize opportunities in large, correlated losses to engage in bad faith practices that delay payouts, deny coverage, or underpay claims (Feinman 2010). As part of the claims adjustment process, homeowners can file disputes and may arrive at negotiated payouts. Following certain disasters, state-run mediation programs provide third-party mediators between policyholders and insurers (Schwarcz 2008), as was the case after Hurricane Sandy (Ballard 2013). Homeowners may also voice their concerns by filing public complaints with state regulatory bodies. Should claims remain unresolved through these pathways, or if no settlement is reached between insurer and insured within the statutory prescriptive period, insureds may sue. Litigation allows insureds to file complaints against their insurance companies for alleged violations of bad faith laws. These statutes, which exist in all states, require insurers to handle claims fairly and promptly, per the Model Unfair Claims Settlement Practices Act (Schwarcz 2008).

In hurricane claims litigation, most discussions of the role of the law in compensating hurricane victims hinge on the questions of fault attribution and standing (Zink 2014). But often, denied insurance claims are blatant acts of neglect and breaches of contract. The egregious nature of denied or delayed claims explains why, in hurricane insurance litigation, legal representation follows a contingency-fee model in which plaintiffs pay lawyers’ fees only if they lose; the business case for contingency-fee models makes sense when the plaintiffs’ representation sees a clear winning case. The often obvious fault of insurers is also why cases settle pretrial, a predictable outcome in insurance litigation (Dulin 2008).

The legal system plays an important role in the governance of disasters (Tierney 2012). Through cases brought and adjudicated, the courts and the legal apparatus address issues that other institutions, firms, and governments deflect (Sterett and Mateczun 2020). Legal scholars argue that judicialization, or the increasing role of courts and judicial processes in resolving political, social, and administrative issues, is particularly apt for addressing social and economic suffering and legal wrongs, including property loss from climate-related hazards (Sterett and Mateczun 2022).

Raced and Classed Participation in Civil Litigation

Scholars have studied the complex social contexts for differences in individual- and group-level proclivities and decisions to engage the law (Ewick and Silbey 1998). Participation, or lack thereof, of affected people in litigation over post-disaster recovery disputes reveals how the legal process and the governance of law can function as exclusionary mechanisms (Nalla et al. 2021). Legal processes can alienate and estrange minority groups during disaster recovery, leading to mistrust of legal remedies (Simington 2023). Restricted and unequal access to justice can lead to significant socioeconomic and racial group differences in familiarity with the law, recognition of the legal system as a resource, and the capacity to exercise legal rights to resolve civil legal disputes (Sandefur 2019). A lack of comprehensive sociodemographic data on parties in civil cases limits understanding of how access to justice differs across geographies and different populations (Sandefur 2008). Yet, it has been widely documented that there are racial, class, and gendered gaps in how people understand and use the law, and in their ability to engage with the legal system (Greene 2015; Legal Services Corporation 2022; Silbey et al. 1993). Access to legal services is a key component in shoring up these gaps in legal access post-disaster (Finger 2015).

Procedural Vulnerability to Frame Disparities in Insurance Litigation

Procedural vulnerability examines how institutions, actors, and systems (re)produce, distribute, contest, and sustain risk (Hsu et al. 2015) and how this is negotiated through everyday struggles (Marino and Faas 2020). In the context of disasters, procedural vulnerability examines how institutions, policies, and plans reflect the social hierarchy imposed by systems of oppression, producing uneven power relations and differential recovery (Rivera et al. 2022). Property insurance has received significant attention for its historic discriminatory roots, disenfranchising non-White homeowners from accessing insurance (Squires et al. 2001) and discriminatory claims-handling practices for low-income minority groups (Baker and McElrath 1996; Bolin and Bolton 1986). Procedural vulnerability sets the expectation that social hierarchies structure the unequal delivery and experience of procedural services, such as implicit biases that shape differential treatment from insurance adjusters or legal services. Here, procedural vulnerability serves as a framing for the ways oppression operates to disadvantage populations who may experience disparate treatment from insurers and discourage or prevent those same populations from taking legal action.

LITIGATION OF INSURANCE CLAIMS FROM HURRICANES LAURA AND DELTA IN SOUTHWEST LOUISIANA

In fall 2020, Hurricanes Laura and Delta caused extensive damage to homes in Calcasieu Parish, a low-lying area about thirty miles inland from the coast in Southwest Louisiana. Hurricane Laura alone severely damaged or destroyed roughly 50 percent of Calcasieu Parish’s housing stock (Community Foundation of Southwest Louisiana 2020). As a result of the extreme losses in Calcasieu Parish, there were practical and logistical challenges in having adjusters inspect properties. Mandatory evacuations were still in effect three weeks after Laura’s landfall (Li et al. 2024). Supply chain shortages of construction materials due to the COVID-19 pandemic inflated construction costs. In Louisiana, a total of 177,691 and 89,451 claims were reported for Hurricanes Laura and Delta, respectively, with $7.4 billion in paid loss for Laura and $750 million for Delta (Louisiana Department of Insurance 2021). Despite these large payouts, slow insurance payouts, and denials of federal government assistance among uninsured households have deferred home repairs. According to an audit of all Louisiana residential property complaints in 2020 and 2021, most delayed claims were due to changes in adjusters assigned to individual claims and lagging communication (Louisiana Legislative Auditor 2022). Other complaints included lower-than-expected claim offers, improper claim denial, and adjuster handling of claims, such as non-thorough or incomplete inspections.

There are numerous stages in loss adjustment, and the handling of claims can differ among insureds. For example, differences can arise in how soon after a declared loss the damaged property is inspected, how thoroughly and fairly the field adjuster inspects the property, how smoothly claims are managed between one or multiple adjusters, how aligned the adjustment is with coverages stated in one’s policy or quoted costs of damage from contractors, or how quickly the claim is settled and a payout is made. Throughout this process of claims adjustment, disagreements or delays can drastically alter how insureds benefit from insurance in terms of repairing damage. Following large disaster events, delays and under-compensation are common, leading to litigation against insurers (Dulin 2008).

At the time of Laura, there were two bad faith statutes in Louisiana, §22:1892 and §22:1973, through which litigants brought claims. Under §22:1892, insurers must make a written offer to settle and pay a claim amount within thirty days of “satisfactory proof of loss” from the insured. Any “arbitrary, capricious, or without probable cause failure” to fulfill this statute results in penalties for the insurer for up to 50 percent of the unpaid amount in addition to the amount of loss and attorney’s fees if the statute is violated. The second statute, §22:1973, further ensures fair practices whereby if an insurer fails to pay claims pursuant to §22:1892, thereby breaching its duty of good faith and fair dealing, penalties up to 200 percent may be imposed. In most cases, the §22:1892 statute applies when settlements are reached before or during mediation, thereby triggering the 50 percent penalties and attorney’s fees.

In 2021, a year after Hurricanes Laura and Delta, the United States District Court for the Western District of Louisiana (LAWD) had the largest number of federal court civil lawsuits of any district in the nation (2,194), 91 percent of which were filed in Lake Charles (TRAC 2021). Within two years of Laura and Delta’s passing, nearly one in five lawsuits filed in federal court were filed in Lake Charles (TRAC 2022). The total reported litigation rates for Laura were 8.33 percent and for Delta, 6.92 percent.

The LAWD quickly became the clearinghouse for all federal cases. Taking cues from the Southern District of Texas Federal Court, which experienced an exponential uptick in cases related to insurance following Hurricane Harvey, Judge James Cain, a federal LAWD judge, created a Case Management Order, which is a court order outlining the procedure for cases on a court’s docket, often setting timelines for discovery. Acknowledging the need to expediently handle the large number of case filings over first-party insurance claims disputes from Hurricanes Laura and Delta, the Case Management Order presents a Streamlined Settlement Process, which provides clear instructions on the timelines for filing defense responses, initial discovery, scheduling pretrial negotiations, and mediation (Western District of Louisiana 2023). Additionally, cases were also filed in the 14th Judicial District Court (14JDC) of Louisiana, which also adopted the Case Management Order enacted and filed in the LAWD. The factors that determine if a case is filed in a state judicial court or a federal court are the monetary amount in question and the diversity of parties; if the disputed claim amount is greater than $75,000 and the plaintiff is a resident of Louisiana but their insurance carrier is domiciled outside the state of Louisiana, these are both criteria for a suit being filed in federal court.1 According to personal correspondence with an attorney in Lake Charles who handled several hurricane claims litigation cases, by early 2024, the majority of the cases had closed with pretrial settlements, a common outcome in insurance litigation of this nature (Sterett and Mateczun 2022).

Insurance is a key concern for Louisianans, and policymakers have pointed to legal system abuse as a key instigator of rising insurance premiums and coverage limits. In response, the insurance industry and industry-aligned legislatures advocate for tort reform, changes in laws that are designed to limit the number of lawsuits brought after large losses and reduce the amount of money insurance companies might have to pay in claims or legal damages.2 Reforms of this type are often introduced after major storms when a surge in insurance disputes and lawsuits can strain insurers and the court system. Proponents of tort reform, notably insurance industry lobbyists and political actors, argue that tort reform stabilizes insurance costs and prevents excessive litigation, while critics argue that it weakens consumer protections. Most of the literature suggests that litigation is valid (Klein 2022).

Despite evidence that tort reform does not lead to reduced premiums (Mower 2023), the 2023–2024 Louisiana State Legislative session has rolled out sweeping changes to how insurance is regulated in the state, including changes to litigation (for example, Landry 2024). The rates of litigation have been debated among state legislatures, industry commentators, legal actors, and consumer advocates, each arriving at different conclusions about the root causes of insurance unaffordability and the role of litigation in moderating insurance costs. Despite growing concerns among residents, consumer advocates, industry affiliates, and policymakers across the nation about rising insurance premiums, industry and tort reform politics routinely label litigation as frivolous and the root cause of those premiums (Jefferson 2024). Under the new changes is Act No. 3 (formerly Senate Bill 323) which repeals La. R.S. 22:1973, thereby removing the 200 percent penalty provision and allowing a longer time for insurers to adjust claims. As these recent legislative changes make it more difficult for people to bring complaints, they raise further questions about who will continue to receive improper claims handling after these new statutes are in place to abate insurance bad faith tactics, and whether the changes will entrench any existing disparities in litigation.

Following the devastating 2020–2021 hurricane seasons, insured policyholders experienced enormous premium increases or experienced cancellations, and approximately fifteen companies left the state entirely (Finch II 2022). Concerns over insurance affordability and access have grown, complicating the rebuilding process. Recovery from Hurricanes Laura and Delta was largely fast-tracked in wealthier, Whiter neighborhoods and delayed in poorer, Black neighborhoods (Maschke 2024). The racialized inequities in recovery in Lake Charles are inherently spatial, with Interstate 10 bifurcating the city into majority-white and majority-Black neighborhoods. Delays in repairs also occurred for uninsured properties, common among heirship properties found within Black households, where inheritors cannot afford to clear title, purchase insurance, or obtain loans to make repairs without insurance (Kane et al. 2019). This differential recovery further motivates interrogation of the sociodemographic pattern of litigation for this study.

HYPOTHESES

Socioeconomic differences in insurance claim experiences, coupled with research showing unequal access in decisions and the ability to invoke the law when there is harm (Ewick and Silbey 1998), create competing expectations about who will pursue litigation. On one hand, minority and lower-income populations may receive poorer treatment from insurers (for example, Flitter 2022), prompting the need for litigation. Yet, race and income play significant stratifying roles in seeding different levels of trust and use of the legal system (Greene 2015). Previous studies examining differences use survey-based data and claims records (for example, Lin et al. 2022), which are often privately held by the insurance industry. However, litigation records alone do not reveal whether socioeconomic dimensions influenced differential claims handling. To determine whether there were disparate experiences with claims adjustment following Hurricanes Laura and Delta would require a systematic sample of insurance claims, which was not provided on request from the Louisiana Department of Insurance. Given our inability to control for disparate experiences with poor insurance claims handling, we cannot assess this as a mechanism in our study.

Instead, we leverage patterns of insurance litigation as evidence of “turning to the law” to resolve disputes. This sample, therefore, may not fully represent all households experiencing poor treatment from their insurance company and does not identify mechanisms that explain any differences in participation. Empirical legal studies demonstrate that affluent, White individuals are more likely than non-White, lower-income individuals to initiate civil litigation and secure legal representation (Myrick et al. 2012). This disparity is driven in part by greater access to legal resources, ease and confidence in navigating the legal system, and the systematic disenfranchisement of minority groups from accessing and trusting the legal system (Simington 2023). This research builds from evidence of legal participation to examine patterns of sociodemographic differences in insurance litigation. We rest our analysis on four hypotheses:

  1. Insurance litigation is more likely in socioeconomically privileged areas.

  2. Plaintiffs in insurance litigation cases are more likely to be residents than nonresidential parties in socioeconomically privileged areas.

  3. Insurance litigation cases are filed earlier in socioeconomically privileged areas.

  4. Insurance litigation cases are resolved more quickly in socioeconomically privileged areas.

DATA

This analysis collates data on litigation at the parcel level along with various other parcel, census block group, and tract-level data for the study area of Calcasieu Parish to test the four hypotheses (table 1).

View this table:
Table 1.

Description of Variables Used to Test Hypotheses

State and Federal Court Records of First-Party Insurance Litigation Cases

Data on state and federally filed first-party insurance lawsuits were collected to create parcel-level data on the incidence of insurance litigation, plaintiff type, and temporal characteristics of case filing dates and durations. For cases filed in the 14JDC, data on first-party insurance lawsuits were provided by the Calcasieu Parish Clerk of Court. The dataset contained 1,998 records and provided the address of the property associated with the litigation, the full name of the plaintiff and defendant, and the date of the case filing. For the cases filed in LAWD, a total of 7,363 first-party insurance lawsuit case dockets were obtained via Public Access to Court Electronic Records. Data on the address of the property associated with the litigation, the full name of the plaintiff and defendant, the date of the case filing, and the date of the case closing were systematically extracted from dockets. Using the addresses in the combined state and federal cases, we geocoded the records and spatially intersected the point locations with the parcel boundary dataset provided by the Calcasieu Parish Police Jury GIS Office. Each parcel boundary is assigned a parcel number (UID), which links to other parcel-level variables described in the following section.

Manual cleaning of the combined litigation records was undertaken, resulting in the removal of 3,883 cases that were either duplicates or irrelevant cases from the federal case set. A first manual check was performed to drop duplicates in both the state and federal case sets. This duplication was expected, as cases were “removed” from the state 14JDC to the federal LAWD, a common practice in which the details of the case, such as the total amount in dispute or the insurer’s out-of-state domiciliary status, would require litigation in federal court. Second, cases were dropped from the federal set if, within the Petition of Damages document that is filed with all cases’ dockets, the nature of the action did not pertain to alleged bad faith actions of the insurer (the “Defendant”) or if complaints were for damages separate from Hurricanes Laura and Delta (for example, an April 2020 hail event). For instance, cases were dropped if they entailed businesses seeking remediation for business interruption for all-risk commercial property insurance policies. Third, records were manually cleaned if the provided address was a post office box, as this did not represent a physical structure. For instances where no known addresses were stated, 2020 tax assessment records were used to locate the property of the plaintiff if only one property was associated with the name of the indicated plaintiff. Lastly, 1,642 cases were removed in which the plaintiff’s lawyer was McClenny Moseley & Associates, as these cases were later dropped following Judge Cain’s order due to fraudulent case management.3 After removing duplicates and cases with addresses outside of Calcasieu Parish, there are 5,476 unique records of litigation participation. The majority of litigation occurs in Lake Charles, the urban area of the Parish (figure 1).

Calcasieu Parish Study Area with Counts of Litigation and Hurricanes Laura and Delta Tracks Source: Authors’ calculations based on recorded litigation from LAWD and JDC14 case dockets. Hurricane tracks are from the National Oceanic and Atmospheric Administration Historical Hurricane Tracks database. Note: The inset map in the upper right shows the study area as located within Louisiana and surrounding states. Each symbolized hexbin is an area of approximately 1.2 square kilometers. Any location without a hexbin within Calcasieu Parish indicates no record of insurance litigation. The optimal way to view this map is in color. We refer readers of the print edition of this article to https://www.rsfjournal.org/content/12/4/178 to view the color version.
Figure 1.

Calcasieu Parish Study Area with Counts of Litigation and Hurricanes Laura and Delta Tracks

Source: Authors’ calculations based on recorded litigation from LAWD and JDC14 case dockets. Hurricane tracks are from the National Oceanic and Atmospheric Administration Historical Hurricane Tracks database.

Note: The inset map in the upper right shows the study area as located within Louisiana and surrounding states. Each symbolized hexbin is an area of approximately 1.2 square kilometers. Any location without a hexbin within Calcasieu Parish indicates no record of insurance litigation. The optimal way to view this map is in color. We refer readers of the print edition of this article to https://www.rsfjournal.org/content/12/4/178 to view the color version.

For each case, the plaintiff type indicator variable was created and classified as either residential or nonresidential. Specifically, any case where the plaintiff type is residential represents a case in which the plaintiff is the residential owner of that property. In nonresidential cases, the plaintiff was a commercial business, a public entity, or another organized entity, such as a church. The plaintiff’s name was used to designate the plaintiff type. If the plaintiff’s name was a single individual (for example, John Smith) or, in some cases, a couple’s names (for example, John and Petra Smith), the plaintiff type was classified as residential. If the terms “Inc.,” “LLC,” “LLP,” or “Co.” were included in the name, they were designated as nonresidential. Remaining records were manually coded to ensure all records where the plaintiff did not appear to be a resident (for example, terms like “Restaurant” or “Church”) were accurately classified as nonresidential. Residential properties accounted for 82 percent of all cases. Resident plaintiffs are distributed across the study area, with higher densities clustered in Lake Charles, whereas nonresidential plaintiffs are mainly located in urban areas (online appendix figure A.1).4

Next, we construct two data variables using the case filing and closing dates. First, we calculate the total time in unit months from Hurricane Laura’s landfall on August 27, 2020, until the date of the filing to create the case filing date variable. Second, for cases closed at the time of data collection in September 2024, we calculate the total time in unit months between the case filing and closing dates to construct the case duration variable. If cases were still ongoing (not closed) at the time of data collection, they were not assigned a case duration value.

Parcel-Level Variables

We collected three variables at the parcel level, relevant for deriving socioeconomic variation in litigation incidence, plaintiff type, and case timing. The first two originate from the 2020 Calcasieu Parish tax assessment records that were obtained from the Louisiana Tax Commission and contain information on the parcel assessment number, the total assessed value of the property, including the lot and property improvements, and whether a property has received a homestead exemption.5 From these records, we first calculate the total property value and, given the pronounced right skew in the data, we top code at the 90th percentile, or $2,272,000. Second, we use the homestead exemption as an indicator for owner-occupied property.

The third parcel-level variable is a time-invariant indicator of insured property indirectly estimated using the CoreLogic Mortgage dataset, which includes property-level data on active mortgages in 2020. Using mortgages as a proxy for insured property is a well-supported approach, as lenders typically require properties with mortgages to be insured, corroborated by a recent study that found 98 percent of mortgaged owner-occupied homes had property insurance (Cornelissen et al. 2024; Keys and Mulder 2024). The insurance indicator controls for insured and uninsured properties with a mortgage that are not involved in insurance lawsuits. Notably, the insurance indicator does not capture insured properties owned outright, which may bias insurance estimates downward. Additionally, this indicator variable does not capture instances of underinsurance, in which insurance coverage limits or the total coverage of policies are less than the property’s total replacement value. Of the total 94,838 parcels, 35,845 (38 percent) were matched to active mortgages at the time of Hurricane Laura’s landfall on August 27, 2020. This figure is reasonable, as not all parcels are residential; they can be commercial, industrial, agricultural, or other zoned land uses. Using the parcel number as the unique identifier, the tax assessment records containing the total assessed value and homestead exemption, along with the mortgage records used to proxy for insured properties, were joined to the parcel layer containing the data on litigation involvement.

Block Group- and Tract-Level Variables

We collect block group- (j) and tract-level (k) data for additional predictors of socioeconomic estimates and hazard impact, which act as groups within which all individual parcels are nested. We assign the parcel-level records that include the litigation and property to census data block groups and tracts, which allows us to merge data from the American Community Survey 2018–2022 Five-Year Estimates. First, we compile block group-level estimates of Hispanic, non-Hispanic White, non-Hispanic Black, non-Hispanic Asian, and non-Hispanic American Indian and Alaska Native (AIAN) populations. At the block group level, we renormalize the group-level percentages by summing across all five race and ethnicity categories. Because the largest racial and ethnic populations within Calcasieu Parish are non-Hispanic White (66 percent) and Black (25 percent), we further renormalize the data to remove the total estimate of Hispanic, non-Hispanic Asian, and non-Hispanic AIAN to calculate the percent non-Hispanic White and non-Hispanic Black populations relative to one another, with the non-Hispanic White later used as the reference in modeling.

We acknowledge trade-offs in using block group-level data with higher margins of error (MOE) at smaller geographic units (Jurjevich 2019). However, because race and ethnicity have a lower relative sampling error given that these data are benchmarked to the 2020 Decennial (US Census Bureau 2024) and because we maximize data heterogeneity at the smaller block group scale, we decide to use the larger variation provided in the block group-level estimates for percent Black since it outweighs the inherent bias present in smaller geographic estimates. For other predictors where the data are not benchmarked to the 2020 Decennial Census and given concerns about large MOE in small population subsets and rural geographies, we use tract-level estimates.

Moreover, the tract-level variables also include population density, median household income, and the percentage of the population with a bachelor’s degree. As with the renormalization procedure for the block group-level variable percent Black, we also calculate it for the tract-level estimate. While there are 161 block groups and 56 total tracts within Calcasieu Parish, 2 block groups and tracts are uninhabited, and 1 block group contains no instances of litigation, resulting in 159 block groups and 54 tracts for which the predictors were compiled. All census tract-level variables are time-invariant. The economic and racially segregated residential geography of Calcasieu Parish is evident in figure 2.

Percent Black Population and Median Property Value in Calcasieu Parish, Louisiana Source: Authors’ calculations based on estimates from the ACS 2018–2022 five-year estimates. Note: Symbolized proportions are displayed for 159 of the 162 census block groups in Calcasieu Parish.
Figure 2.

Percent Black Population and Median Property Value in Calcasieu Parish, Louisiana

Source: Authors’ calculations based on estimates from the ACS 2018–2022 five-year estimates.

Note: Symbolized proportions are displayed for 159 of the 162 census block groups in Calcasieu Parish.

Additionally, the joint National Institute of Standards and Technology–Applied Research Associates, Inc. rapid-response wind contours are used to generate tract-level maximum peak wind speeds (mph) for both Hurricanes Laura and Delta (NIST/ARA 2020). We disaggregate vector-based wind contours to the tract level. While extreme winds and rain characterized both Hurricanes Laura and Delta, we chose to include only the maximum wind speed for Hurricane Laura, as it was predominantly a wind-driven event, whereas Delta was a rain-driven event (Cangialosi and Berg 2021; Pasche et al. 2021). Of the two hurricanes, Laura wrought the most damage to structures; therefore, estimates of maximum peak wind speed are used to control for geographic variation in hazard magnitude. The decision to include wind data is motivated by the need to control for the exogenous and heterogeneous impact of the hazard when statistically modeling variation in disaster outcomes (Kousky 2019). We use wind speed to account for variation in hazard, as there is no publicly available data to control for property-level damage. We caution against using damage estimates, given the possible endogeneity and omitted-variable bias in underlying population and housing characteristics that determine both damage and litigation participation, which could therefore bias estimates (Bakkensen and Larson 2014).

Descriptive Statistics

Summary statistics showcase the differences in litigation across parcels, as well as the tract-level sociodemographic variables. First, 5,476 (about 6 percent) parcels in Calcasieu Parish filed property claims lawsuits following Hurricanes Laura and Delta (table 2). The median assessed property value of households filing claims was $115,200 compared to $53,100 for parcels not involved with litigation. Of the litigated parcels, 66 percent were owner-occupied. These statistics highlight that litigation entails mainly private, owner-occupied residents suing their insurance company. This high proportion of owner-occupied properties for parcels with litigation tracks with the 82 percent of plaintiff types classified as “residential” as compared to 18 percent as “nonresidential.” The difference between the percent of owner-occupied parcels and those assigned “residential” plaintiff type is likely because not all residents claim the homestead exemption.

View this table:
Table 2.

Summary Statistics for Parcel-Level Records

Defendants are not presented in the summary statistics; however, State Farm and Allstate are involved in the most cases, accounting for nearly 40 percent of defendants in insurance litigation cases. While the filing or opening and closing dates are available for federal cases, only the filing dates are provided for state cases. Therefore, the summary statistics for the filing date pertain to both state and federal cases filed with the state JDC14 and the federal LAWD court, while the summary statistics for the closing date include only cases filed with LAWD. The most frequent filing date is August 25, 2022, immediately before the two-year prescriptive period. The median duration from filing to closure is thirteen months, with a maximum of forty-three months, indicating the case was still open at the time of data collection in September 2024 (table A.1). Table A.2 presents summary statistics for the block group- and tract-level variables. There is no strong indication of correlations among the variables, with a moderate negative correlation seen among percent Black and income (figure A.2).

EMPIRICAL APPROACH

We use general linear mixed models (GLMMs) to test all four hypotheses. In all models, the dependent variable is parcel-level, the independent variables are parcel-level (i), block group-level (j), and tract-level (k), and depending on the hypothesis being tested, the total N varies (table 3). Mixed models are well-suited to handle the hierarchical structure of a parcel-level dependent variable and a combination of parcel-, block group, and tract-level predictors with the inclusion of fixed and random effects. All GLMM modeling is conducted using the lme4 package in R (Bates 2015). We use a binary logistic GLMM to assess socioeconomic predictors of the probability of insurance litigation (model 1) and plaintiff type (model 2) and a Poisson GLMM to estimate case filing date (model 3) and case duration (model 4). We use a Poisson count model to estimate case filing date and duration as both represent counts in months and are nonlinear as they do not take on negative values, and are censored since not all cases were closed at time of analysis.

View this table:
Table 3.

Models and Variables Used to Test Hypotheses

For model convergence, z-scores are produced for the parcel-level top-coded property value variable and the tract-level variables of population density, income, percent Bachelor’s, and Hurricane Laura maximum wind speed to reduce variable skew. Across models 1–4, we specify a baseline model and a model with two interactions. In the baseline specification (models 1–4A), there are eight independent variables in the model, which include the parcel-level property value, the owner-occupied indicator, the insured indicator, the block group-level percent non-Hispanic Black, and tract-level population density, income, percent Bachelor’s, and Laura maximum wind speed. Interactions are also performed between percent Black and top-coded property value (models 1–4B) and percent black and income (models 1–4C) to evaluate the direction and magnitude of the association between the various outcomes and socioeconomic privilege. In models 3 and 4, an additional interaction between plaintiff type and percent black is added (models 3–4D). Coefficients for model 3–4 estimates are presented as incident rate ratios, which represent the multiplicative change in the expected count of months given a one-unit increase in each predictor. Because they are ratios, they signify relative change rather than absolute change in time units of months. To avoid multicollinearity, the variable inflation factor (VIF) is calculated for each model to ensure that no predictors are multicollinear, with the square root of the VIF not exceeding 2 (Fox 2020).

Across all models, random effects are included at the block group and tract levels. With the random effects in GLMMs, we adjust for group-level correlation by allowing the intercepts to vary across groups. With the GLMM specifications, we were unable to obtain robust clustered standard errors, a challenge with GLMM models. While coefficients would remain the same, standard errors may be underestimated, and significance may therefore be overstated. To address some of these concerns, we apply numerous robustness checks to reestimate models and test the sensitivity of our results. Moreover, even if clustered robust standard errors were obtained, they may not substantively alter coefficient magnitudes, as there is no clustering in the sampling design or in the assignment of effects in our empirical design (Abadie et al. 2023). We then perform model diagnostics by computing the log-likelihood.

Tests for Spatial Autocorrelation

Given the spatial character of the data, we test for spatial autocorrelation to account for any spatial dependency in the outcome, which, if left uncontrolled, can bias standard errors (Anselin 2003). For models 1 and 2, where the dependent variable is binary, the joint-count statistics test for spatial autocorrelation, and in models 3 and 4, where the outcome is numeric, we use Moran’s I statistic to test for spatial autocorrelation. The spatial lag for all models is calculated using the dependent variable for each parcel centroid and an inverse distance weighted matrix of the four nearest neighboring parcel centroids. The spdep package in R is used to construct the spatial lag and test for dependence in the model residuals, first without the model’s associated spatial lag and later with the spatial lag. If autocorrelation is present, signaling that litigation incidence (H1), plaintiff type (H2), time of litigation filing (H3), or case duration (H4) are spatially clustered, a spatial lag is included in the respective models.

Robustness Checks

We performed four robustness checks to ensure model results hold across different data subsets and variable definitions. If spatial autocorrelation was detected in the model without the spatial lag, prompting the inclusion of the spatial lag in the models, the first robustness check is to ensure model results hold across different data subsets and variable definitions without the spatial lag (models 1–4E). Second, because the proxy for insured properties is inferred indirectly from mortgage records, we run all baseline models without the insured variable (models 1–4F). Third, given the trade-offs discussed in using block-group versus tract-level estimates for the racial and ethnic composition variables, there are trade-offs in data scale and potential underlying error in the estimates. Given the trade-offs of using block groups, which capture larger variability at the cost of a high margin of error, we also specify all baseline models using percent Black at the tract-level to determine whether results hold with this lower-resolution variable (models 1–4G). Lastly, while we control for population density at the block-group or tract level, we perform a robustness check on samples within urban areas, as there may be an inherent bias toward litigation due to proximity to lawyers’ offices, which may facilitate easier access and a higher proclivity to litigate. To check whether results hold in the face of this geographic factor, we use Census Place boundaries (figure A.3), which encompass higher-populated areas, to restrict the total sample across all four models (models 1–4H).

RESULTS

All models exhibited significant positive spatial autocorrelation, indicating spatial dependency. Therefore, the spatial lag terms were included in all models, except where noted for robustness checks.

Estimates of Litigation Probability

Parcel-level predictors showed that higher-valued, owner-occupied, and insured properties are significantly more likely to be involved in litigation (table A.3). Notably, areas with higher percentages of Black residents and higher population density also showed significantly increased litigation rates. In the baseline scenario, the odds of litigation increase by 43 percent in an all-Black populated area, suggesting that the probability of litigation increases with the percent Black population. Income had a negligible and statistically insignificant effect on litigation probability.

The interaction between property value and percent of Black residents is positive and significant (figure 3, panel A). For each standard deviation increase in property value (z-score), the odds of litigation increase by 13 percent. This suggests that higher property values in areas with more Black residents may be associated with a higher probability of litigation compared to areas with fewer Black residents (figure 4, panel B). However, because this interaction is smaller than both main effects, it suggests that the joint impact of high property value and high percent Black is not as pronounced as their individual contributions alone might suggest. The interaction between income and percent of Black residents was negative, indicating that the positive relationship between Black population and litigation was weaker in higher-income areas.

Sociodemographic and Property-Related Estimates and Predicted Probability of Property Insurance Litigation Source: Authors’ calculations based on model estimates. Note: N = 94,820
Figure 3.

Sociodemographic and Property-Related Estimates and Predicted Probability of Property Insurance Litigation

Source: Authors’ calculations based on model estimates.

Note: N = 94,820

Sociodemographic and Property-Related Estimates and Predicted Probability of Plaintiff Type Residential Source: Authors’ calculations based on model estimates. Note: N = 5,476. The optimal way to view this figure is in color. We refer readers of the print edition of this article to https://www.rsfjournal.org/content/12/4/178 to view the color version.
Figure 4.

Sociodemographic and Property-Related Estimates and Predicted Probability of Plaintiff Type Residential

Source: Authors’ calculations based on model estimates.

Note: N = 5,476. The optimal way to view this figure is in color. We refer readers of the print edition of this article to https://www.rsfjournal.org/content/12/4/178 to view the color version.

The differences in the estimated odds of litigation between the two interaction terms can be better understood by examining the relationship between property value and income. While these variables exhibit a weak positive correlation (figure A.2) and binned median income generally increases with property value (figure A.4), this pattern is not uniform across the study area. Geographic mismatches exist, with regions showing high aggregated median property value not always aligning with areas of high income by tract (figure A.5). Estimates remain stable across robustness checks (table A.7). In addressing hypothesis 1, we find that litigation participation is less likely in socioeconomically privileged areas by race and ethnicity, but that property value strongly increases odds of litigation. This suggests that property value is a characteristic of litigation across sociodemographics and an expression of privilege in insurance litigation.

Estimates of Plaintiff Type

Estimates from model 2 show that parcel-level property value is a negative and significant predictor of plaintiffs in which the litigant is a resident rather than a nonresident (figure 4, panel A; table A.4). The estimates for the percent Black population are negative but insignificant. Interestingly, the interaction between property value and percent Black remains negative and significant, suggesting that as property value increases, the likelihood of being a resident plaintiff decreases, and that this effect is even more pronounced in higher percent Black areas (figure 4, panel B). Importantly, income is a positive and significant predictor of residential plaintiffs. The distribution of property values for nonresident plaintiffs was higher than for resident plaintiffs (figure A.6, panel A), likely due to higher valued commercial properties. Resident plaintiffs were more concentrated in higher-income areas (figure A.6, panel B). These findings help disentangle litigation’s unique economic and racial geography for different actors. In testing hypothesis 2, we find that plaintiffs in insurance litigation cases are more likely to be residents than nonresidents in socioeconomically privileged areas, but this effect is evident only through income, not race or ethnicity.

The insured indicator is estimated as a negative and significant predictor of residential plaintiffs, potentially due to underestimates of insured properties. The percent Black population is a negative and insignificant predictor of resident plaintiffs. Additionally, population density is a negative and weakly significant predictor of resident plaintiffs, suggesting that more rural areas are more likely to contain resident plaintiffs than nonresidential plaintiffs (figure A.1). Findings for estimating plaintiff type also remain consistent across the four robustness checks (table A.8).

Estimates of Case Filing Date

Model results predicting when lawsuits were filed suggest that owner-occupied properties result in later filing dates by 2 percent, and for insured properties, time to case filing decreases by 5 percent (table A.5). While both are significant, these effects are small. Tract-level variables show little and insignificant variation. Robustness checks suggest these patterns remain stable for the baseline model (table A.9). Predominantly Black areas saw earlier filings overall (11 percent), but this trend reversed for resident plaintiffs, who filed 17 percent later than nonresident plaintiffs (model 3D). This suggests that nonresidential plaintiffs, such as commercial businesses or churches, are more likely to file lawsuits more quickly in majority-Black areas than individual residents are.

Figure A.7 validates these findings, where the mean filing date of residential plaintiffs is slightly higher than that of nonresidential plaintiffs, even though both groups exhibited a common peak in filings at the twenty-four-month mark. However, nonresidents tended to file under the earlier twelve-month deadline, whereas residents filed under the twenty-four-month deadline. Again, these differences are relative given the significantly larger number of resident plaintiffs (4,496) compared to nonresidential plaintiffs (980). These model estimates suggest that hypothesis 3 is unsupported in that insurance litigation cases are filed at relatively the same point in time following Hurricane Laura, regardless of sociodemographics.

Estimates of Case Duration

Estimates of case duration suggest case duration is 8 percent shorter for owner-occupied properties in closed cases (4,776 of the total 5,476) (see figure 5, panel A; table A.6). Given the relationship between owner-occupied properties and the plaintiff type, resident plaintiffs’ cases are 14 percent shorter than nonresident cases (model 4D). Areas with higher percentages of bachelor’s degrees had a modestly significant increase in case duration (3 percent). The interaction between property value and the percent Black population suggests a significant 6 percent increase in case duration (model 4B). Robustness checks suggest these patterns remain stable for the baseline model (table A.10). Model 4D reveals that resident plaintiffs and those living in predominantly Black areas experience shorter case durations but also see that the percent Black populations lead to a 26 percent increase in case duration for resident plaintiffs when accounting for plaintiff type (figure 5, panel B). This suggests that while cases in Black-majority areas tend to resolve faster overall, resident plaintiffs in these areas face longer case durations. We conclude that hypothesis 4 is supported in that the duration of litigation is less in socio-economically privileged areas by race and ethnicity when controlling for plaintiff type.

Sociodemographic and Property-Related Estimates and Predicted Probability of Case Duration Source: Authors’ calculations. Note: N = 4,776
Figure 5.

Sociodemographic and Property-Related Estimates and Predicted Probability of Case Duration

Source: Authors’ calculations.

Note: N = 4,776

DISCUSSION

Systemic racial and class variation explain differential treatment in property insurance claims handling (Lin et al. 2022; Peacock and Girard 1997), the administration and delivery of government assistance post-disaster (Emrich et al. 2022), and recovery outcomes (Muñoz and Tate 2016). The impacts of differential administration in aid and quality of insurance are long-lasting and can be detrimental to aspects that extend beyond household finance, including wealth generation (Howell and Elliot 2019), residential mobility (Rhodes and Besbris 2022), and displacement (Fussell and Harris 2014). To further describe how inequality is perpetuated in disasters, this analysis makes a first foray into describing socioeconomic trends in legal claims against private insurers that are made in post-disaster settings. Our findings echo the evidence of socioeconomic disparities documented in other research on post-disaster recovery. In doing so, this work contributes to the growing empirical efforts to document outcomes stemming from potentially procedural inequalities in disaster-impacted communities (Raker 2023).

These findings clarify the who, where, and when of insurance litigation in the case of Southwest Louisiana following Hurricanes Laura and Delta, suggesting both race and class disparities exist in explaining who brings litigation against insurers. Notably, counter to prevailing expectations of who may be best situated to bring lawsuits, litigation is more likely in higher percent Black residential areas, adding nuance to existing qualitative work finding that minority populations are often disincentivized from participating in legal recourse for harms, particularly in post-disaster recovery contexts (Nalla et al. 2021; Simington 2023). This finding also empirically advances understanding of the socioeconomic gaps in accessing justice through civil litigation procedures (Legal Services Corporation 2022).

Importantly, we do not control for direct property damage, and we lack data on which properties were associated with property insurance claims that closed without litigation, or any individual- or group-level measure of willingness to litigate. Without controlling for these, we are unable to draw any causal linkages between blatant, systematic, race- and class-based discriminatory practices by insurance companies in adjusting claims, which may explain the patterns of higher litigation evident in the findings. Instead, our results only provide evidence of patterns of socioeconomic differences in litigation participation, and do not discern instances in which litigation was unnecessary or pursued due to unknown constraints. As such, our findings in estimating litigation represent just the tip of the grievance-claims-disputes pyramid that Richard E. Miller and Austin Sarat (1981) present as a logic for explaining why not all harms eventually make their way to formal legal claims or disputes. In other words, the absence of litigation may not reflect fair claims handling but rather cases in which individuals were not empowered to sue their insurer or were resource-constrained for various financial or immaterial reasons. In these instances, the need for immediate cash relief to start repairs or hire a contractor outweighed the potentially long and uncertain path of engaging in litigation. While measuring these initial differences in claims handling is not possible with the available data, our analytical design provides evidence of these differences in participation. Further ongoing work by the lead author uses qualitative interviews with residents to understand individuals’ motivations for participating in litigation, illuminating whether and to what extent social and economic factors influence one’s proclivity or ability to engage in litigation following Hurricanes Laura and Delta.

Our findings on socioeconomic differences in the case filing and case duration also reveal unique patterns of the temporality of litigation, further enhancing our understanding of the procedural character of litigation. Our findings that nonresident plaintiffs file cases earlier than residents may suggest that commercial plaintiffs get their foot in the door with court proceedings earlier than residents do, and that businesses may be more legally savvy than residents. Moreover, our finding that case duration is longer in higher percent Black areas may indicate systemic delays for individual homeowners compared to businesses in majority-Black areas. Ongoing research by the lead author to further clarify the mechanisms that drive individuals to engage in litigation against their insurer and survey the array of experiences with litigation to explore how litigation enables recovery aims to elucidate these trends further.

The highly significant spatial clustering observed across all outcome variables suggests that litigation incidence, litigant characteristics, and temporal patterns explored in this analysis exhibit strong spatial dependence. This phenomenon of litigation clustering may be explained by socialized knowledge transfer, as found in other studies to drive awareness of legal remedies (Sandefur 2012), such that residents learn about litigation from neighbors and nearby residents. This spreading of information is consistent with other studies of information spread and knowledge transfer in post-disaster contexts (Aldrich 2014) and the efficacy of shared narratives from collective experiences (Chamlee-Wright and Storr 2011).

These findings are instructive for several possible extensions that could refine the characterization of disparities in insurance litigation. Given that homeowner’s insurance is regulated by states in the US, laws regarding grounds for litigation against insurers post-disaster, including the maximum allowed penalties, the prescriptive period for bringing a suit, and entities exempt from liability for property insurance claims, vary across states. The recent state legislative changes in Florida and Louisiana to deter litigation show how the legal environment and conditions vary across jurisdictional and temporal contexts. Similarly, third-party mediation programs may alleviate the need for formal litigation. Moreover, further analysis can be done to evaluate the degree to which losses incurred following a disaster correlate with litigation rates. Highly spatially correlated claims may exacerbate the need for homeowners to pursue litigation. These variations provide fertile ground for comparative statistical analyses to examine how litigation rates and sociodemographics vary across geographic, legal, and disaster-impact settings. For future survey-based work to extend these findings to other case studies, the results could inform sampling using modeled characteristics of populations likely to engage in litigation, as well as survey-based post-stratification weighting using socioeconomic trends.

The two main limitations of our approach are variable measurement and geographic scale. Regarding variable measurement, we include a robustness check that excludes the proxy for insured properties. Evidence from estimating plaintiff type also suggests possible undermeasurement of insurance, indicating that this proxy is an imperfect measure of properties with homeowners’ insurance. While litigation is predicated on a property being insured, a more precise measure of insurance coverage, along with measures of owner occupation, may help illuminate circumstances in which litigation pertains to residential but non-owner-occupied housing. In this analysis, the zero indicator for non-occupied properties includes rental properties but also applies to nonresidential properties. Given the overwhelming emphasis on researching owner-occupied housing, leaving post-disaster dynamics for renters relatively opaque (Lee and Van Zandt 2019), this analysis further reinforces this focus on owner dynamics, which primarily concern owner-occupied residents; there are possible extensions to clarify how litigation impacts rental property and tenants. To do so, the names of plaintiff types could be further examined to separately classify rental properties from non-rental properties, as well as other commercial or organization-owned properties. Moreover, this would also have to be done with the full parcel sample. Additional data on rental properties or large-scale survey work would need to be undertaken to capture this dynamic, especially how litigation concerning rental properties relates to any disparate impact of displacing tenants while repairs remain partially or wholly unaddressed.

Additionally, omitted variables may influence variable measurement and model specification. Calcasieu Parish residents’ prior experiences with hurricane-related property damage, particularly Hurricane Rita in 2005, may affect their expectations and familiarity with navigating property insurance claims for hurricane losses. Law clinics have been proven essential sources of education and legal aid for residents, particularly low-income residents who may have the financial resources to hire a lawyer but need assistance with insurance claims adjustments and other legal needs that arise post-disaster (Finger 2015). The degree or character of these interventions’ influence on residents’ litigation participation cannot be discerned from the data available for this analysis. Relatedly, the extent to which fraudulent lawyering biases the measurement of litigation participation is beyond the scope of this study. However, we remove all McClenny Moseley & Associates cases, which represent the bulk of reported fraudulent lawyering.

The second main limitation is the scale of the estimates and predictors. Who these litigants are in terms of sociodemographics says a lot about the geography of litigation, given how segregated the Parish is racially and economically, except for tracts located centrally in Lake Charles. Importantly, though, we caution against drawing ecological inference in assuming that all Black residents in relatively high Black percent population block groups litigate or that only residents living in those tracts are Black. Moreover, given the spatial scale of the socioeconomic data, we are limited in drawing strong conclusions about the geographic pattern of litigation, as it maps to areas with diverse racial and economic residential makeup. For example, in tracts with a relatively balanced mixed proportion of Black and White residents in central Lake Charles, there is a high rate of litigation, obscuring precise estimates of litigants’ socioeconomic characteristics. While there are methods for imputing race at the individual level (see Hepburn et al. 2020), we chose not to do so, given both empirical and ethical concerns about the validity of the estimates. Further work to refine the racial estimates also risks overfitting the model and hindering generalization to other contexts. While we chose not to pursue individual race imputation given these justifications, we can only conclude neighborhood sociodemographic variations in who brings suits, when they file, and how long their cases last at the block-group and tract levels.

CONCLUSION

While the decision to litigate ultimately hinges on the insured picking up the phone to call a lawyer, this act of turning to the law to resolve unmet damages is prompted by a series of decisions, actors, and social encounters, each shaping one’s conceptualization of the worth of litigation and ultimately implicating them to pragmatically act in mobilizing the law (Blomley 2014). This complex landscape of litigants and experience mirrors what Kate Booth and Bruce Trantor (2017) describe as “temporarily shifting and heterogeneous spaces of trust appear to embody significant and differentiated power relations, especially when it comes to the relationship between households and insurance companies.” In individual insurance cases, the legal claims signal not only alleged misconduct by their insurer but also a decision made at some point in time to bring a lawsuit.

This study documents and analyzes the socioeconomic and housing characteristics of recorded litigation cases. Existing summaries of insurance litigation provide non-specific descriptions of litigation proceedings and parties involved, masking who brings suits against insurers. This risks assumptions about the role litigation plays in recovery, given procedural disparities in how different groups navigate and what they potentially gain or lose from insurance litigation. Here, we present a novel undertaking to collate court records on insurance litigation, geolocated to the property level, to infer the spatial disparities in legal claims-making in post-disaster contexts, a significantly understudied phenomenon. Evaluation of this dataset with sociodemographic estimates within Calcasieu Parish yields insights that, in this case setting, Black residents litigate at a higher rate than White residents, with property value acting as a key determinant of litigation. While our approach precludes assessing the mechanisms of litigation and potentially race- or class-based discrimination in claims handling, it offers an empirical approach to expose patterns of litigation, which are fruitful points of departure to funnel empirical attention to the particular social, economic, and policy contexts that necessitate litigation and implications of state reforms to clamp down on litigation.

This article interrogates and argues that insurance litigation is yet another facet through which procedural vulnerability is used to understand the institutions, actors, and systems that (re)produce, distribute, contest, and sustain risk and alternative futures (Marino and Faas 2020). As disasters become more frequent and severe, this work joins debates on how individuals navigate everyday risks and procedural challenges through insurance and litigation.

FOOTNOTES

  • ↵1. According to the Louisiana Department of Insurance, the average severity per claim in litigation is $99,500 versus $22,500 for non-litigated claims.

  • ↵2. Common tort reform measures include caps on the amount of money policyholders can recover, stricter filing deadlines after the event, restrictions on how much lawyers can earn from insurance lawsuits, which can make it harder for policyholders to find legal representation, and altering “bad-faith” laws. These changes can make it harder for policyholders to find legal representation and raise the bar for policyholders to sue insurers for acting unfairly or denying claims improperly. In the 2024 State Legislative Session, Senate Bill 113 was signed into law, which deems Louisiana Citizens Property Insurance Cooperation, the state insurer of last resort, exempt from litigation.

  • ↵3. It is beyond the scope of this article, but the McClenny Moseley & Associates court order has embroiled and fueled political arguments that all instances of insurance litigation are frivolous.

  • ↵4. The online appendix for tables A.1 through A.10 and for figures A.1 through A.7 can be found at https://www.rsfjournal.org/content/12/4/178/tab-supplemental.

  • ↵5. In the state of Louisiana, the homestead exemption is an exemption from state and parish taxes up to $7,500 of the total assessed value for primary residences where the property is owner-occupied.

Open Access Policy: RSF: The Russell Sage Foundation Journal of the Social Sciences is an open access journal. This article is published under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License.

REFERENCES

  1. ↵
    Abadie, Alberto, Susan Athey, Guido W. Imbens, and Jeffrey M. Wooldridge. 2023. “When Should You Adjust Standard Errors for Clustering?” Quarterly Journal of Economics 138(1): 1–35. https://doi.org/10.1093/qje/qjac038.
    OpenURLCrossRef
  2. ↵
    Aldrich, Daniel P., and Michelle A. Meyer. 2014. “Social Capital and Community Resilience.” American Behavioral Scientist 59(2): 254–69. https://doi.org/10.1177/0002764214550299.
    OpenURL
  3. ↵
    Anselin, Luc. 2003. “Spatial Econometrics.” In A Companion to Theoretical Econometrics, edited by Badi H. Baltagi. Blackwell Publishing Ltd.
    Google Scholar
  4. ↵
    Baker, Tom, and Karen McElrath. 1996. “Whose Safety Net? Home Insurance and Inequality.” Law & Social Inquiry 21(2): 229–64. https://doi.org/10.1111/j.1747-4469.1996.tb00080.x.
    OpenURL
  5. ↵
    Bakkensen, Laura A., and William Larson. 2014. “Population Matters When Modeling Hurricane Fatalities.” Proceedings of the National Academy of Sciences 111(50): e5331–e5332. https://doi.org/10.1073/pnas.1417030111.
    OpenURLPubMed
  6. ↵
    Ballard, Jordan, Julia Howard-Gibbon, Branda Muñoz Furnish, and Aaron Scheinwald. 2013. “Natural Disasters, Access to Justice, and Legal Services.” City University of New York Law Review 17(1). https://doi.org/10.31641/clr170101.
  7. ↵
    Bates, Douglas, Martin Mächler, Ben Bolker, and Steve Walker. 2015. “Fitting Linear Mixed-Effects Models Using lme4.” Journal of Statistical Software 67(1): 1–48. https://doi.org/10.18637/jss.v067.i01.
    OpenURL
  8. ↵
    Blomley, Nicholas. 2014. “Learning from Larry: Pragmatism and the Habits of Legal Space.” In The Expanding Spaces of Law: A Timely Legal Geography, edited by Irus Braverman, Nicholas Blomley, David Delaney, and Alexandre Kedar. Stanford University Press.
    Google Scholar
  9. ↵
    Bolin, Robert C., and Patricia A. Bolton. 1986. “Race, Religion, and Ethnicity in Disaster Recovery.” Institute of Behavioral Science, University of Colorado. January. https://scholarcommons.usf.edu/fmhi_pub/88.
  10. ↵
    Booth, Kate, and Bruce Tranter. 2017. “When Disaster Strikes: Under-Insurance in Australian Households.” Urban Studies 55(14): 3135–50. https://doi.org/10.1177/0042098017736257.
    OpenURL
  11. ↵
    Cangialosi, John P., and Robbie Berg. 2021. Hurrican Delta (AL262020): Tropical Cyclone Report. National Hurricane Center. National Oceanic and Atmospheric Administration. April 19. https://www.nhc.noaa.gov/data/tcr/AL262020_Delta.pdf.
  12. ↵
    Chamlee-Wright, Emily, and Virgil Henry Storr. 2011. “Social Capital as Collective Narratives and Post-Disaster Community Recovery.” Sociological Review 59(2): 266–82. https://doi.org/10.1111/j.1467-954X.2011.02008.x.
    OpenURL
  13. ↵
    Community Foundation of Southwest Louisiana. 2020. “Hurricanes Laura & Delta Recovery.” Community Foundation of Southwest Louisiana. November. https://static1.squarespace.com/static/5b6c60224611a06e12b5da29/t/600f4c5cd52c160bdea1a735/1611615327310/202011%2BHurricane%2BLaura%2BDelta%2Brecovery_vFINAL.pdf.
  14. ↵
    Cornelissen, Sharon, Douglas Heller, and Michael DeLong. 2024. Exposed: A Report on 1.6 Trillion Dollars of Uninsured American Homes. Consumer Federation of America. March 12. https://consumerfed.org/wp-content/uploads/2024/03/Exposed-UninsuredHomes-1.pdf.
  15. ↵
    Dulin, Kelsey D. 2008. “The Disaster After the Disaster: Insurance Companies’ Post-Catastrophe Claims Handling Practices.” Oklahoma Law Review 61(1): 189.
    Google ScholarOpenURL
  16. ↵
    Emrich, Christopher T., Sanam K. Aksha, and Yao Zhou. 2022. “Assessing Distributive Inequities in FEMA’s Disaster Recovery Assistance Fund Allocation.” International Journal of Disaster Risk Reduction 74: 102855. https://doi.org/10.1016/j.ijdrr.2022.102855.
    OpenURL
  17. ↵
    Ewick, Patricia, and Susan S. Silbey. 1998. The Common Place of Law: Stories from Everyday Life. University of Chicago Press.
    Google Scholar
  18. ↵
    Feinman, Jay M. 2010. Delay, Deny, Defend: Why Insurance Companies Don’t Pay Claims and What You Can Do About It. Delden Press.
    Google Scholar
  19. ↵
    Finch II, Michael. 2022. “Here Are the Louisiana Insurers That Have Gone Broke or Left the State amid Deepening Crisis.” Nola.com, September 27. https://www.nola.com/news/business/here-are-the-louisiana-insurers-that-have-gone-broke-or-left-the-state-amid-deepening/article_c7f077b4-3e98-11ed-86c9-f7f11037202f.html.
  20. ↵
    Finger, Davida. 2015. “Post-Disaster Housing Through the Lens of Litigation: The Katrina Housing Justice Docket.” Loyola Law Review 61: 591–621.
    Google ScholarOpenURL
  21. ↵
    Finger, Davida. 2022. “50 Years After the ‘War on Poverty’: Evaluating the Justice Gap in the Post-Disaster Context.” Boston College Journal of Law & Social Justice 34(2): 267–82.
    Google ScholarOpenURL
  22. ↵
    Flitter, Emily. 2022. “New Suit Uses Data to Back Racial Bias Claims Against State Farm.” New York Times, December 14. https://www.nytimes.com/2022/12/14/business/state-farm-racial-bias-lawsuit.html.
  23. ↵
    Fox, John David. 2020. Regression Diagnostics: An Introduction, 2nd ed. SAGE Publications, Inc. https://doi.org/10.4135/9781071878651.
  24. ↵
    Friedman, Lawrence M. 1975. The Legal System: A Social Science Perspective. Russell Sage Foundation.
    Google Scholar
  25. ↵
    Fussell, Elizabeth. 2015. “The Long-Term Recovery of New Orleans’ Population After Hurricane Katrina.” American Behavioral Scientist 59(10): 1231–45. https://doi.org/10.1177/0002764215591181.
    OpenURLCrossRefPubMed
  26. ↵
    Fussell, Elizabeth, and Elizabeth Harris. 2014. “Homeownership and Housing Displacement After Hurricane Katrina Among Low-Income African-American Mothers in New Orleans.” Social Science Quarterly 95(4): 1086–100.
    Google ScholarOpenURLPubMed
  27. ↵
    Greene, Sara Sternberg. 2015. “Race, Class, and Access to Civil Justice.” Iowa Law Review 101(4): 1263–322.
    Google ScholarOpenURL
  28. ↵
    Hepburn, Peter, Renee Louis, and Matthew Desmond. 2020. “Racial and Gender Disparities Among Evicted Americans.” Sociological Science 7(27): 649–62. https://doi.org/10.15195/v7.a27.
    OpenURL
  29. ↵
    Howell, Junia, and James R. Elliott. 2019. “Damages Done: The Longitudinal Impacts of Natural Hazards on Wealth Inequality in the United States.” Social Problems 66(3): 448–67. https://doi.org/10.1093/socpro/spy016.
    OpenURL
  30. ↵
    Hsu, Minna, Richard Howitt, and Fiona Miller. 2015. “Procedural Vulnerability and Institutional Capacity Deficits in Post-Disaster Recovery and Reconstruction: Insights from Wutai Rukai Experiences of Typhoon Morakot.” Human Organization 74(4): 308–18.
    Google ScholarOpenURL
  31. ↵
    Jefferson, Elana Ashanti. 2024. “Triple-I CEO Links Catastrophe Costs to Legal System Abuse.” PropertyCasualty360, July 11. https://www.propertycasualty360.com/2024/07/11/triple-i-ceo-links-catastrophe-costs-to-legal-system-abuse/.
  32. ↵
    Johnson, Leigh. 2015. “Catastrophic Fixes: Cyclical Devaluation and Accumulation Through Climate Change Impacts.” Environment and Planning A: Economy and Space 47(12): 2503–21. https://doi.org/10.1177/0308518x15594800.
    OpenURL
  33. ↵
    Jurjevich, Jason R. 2019. “Confronting Statistical Uncertainty in Rural America: Toward More Certain Data-Driven Policymaking Using American Community Survey (ACS) Data.” In Population, Place, and Spatial Interaction, edited by Rachel S. Franklin. New Frontiers in Regional Science: Asian Perspectives, vol. 40. Springer Nature Singapore.
    Google Scholar
  34. ↵
    Kane, Christy, Stephanie Beaugh, and Gerren Sias. 2019. “Addressing Heirs’ Property in Louisiana: Lessons Learned, Post-Disaster.” In Heirs’ Property and Land Fractionation: Fostering Stable Ownership to Prevent Land Loss and Abandonment, edited by Cassandra J. Gaither, Ann Carpenter, Tracy Lloyd McCurty, and Sara Toering. US Department of Agriculture Forest Service, Southern Research Station.
    Google Scholar
  35. ↵
    Keys, Benjamin J, and Philip Mulder. 2024. “Property Insurance and Disaster Risk: New Evidence from Mortgage Escrow Data.” NBER Working Paper 32579. National Bureau of Economic Research. https://www.nber.org/system/files/working_papers/w32579/w32579.pdf.
  36. ↵
    Klein, Ken. 2022. “Unpacking ‘Social Inflation.’” NAIC Consumer Liaison Committee, NAIC Summer 2022 National Meeting. August 12. https://content.naic.org/sites/default/files/national_meeting/AttmtFive_Consumer_Social%20Inflation_kenklein.pdf.
  37. ↵
    Kousky, Carolyn. 2019. “The Role of Natural Disaster Insurance in Recovery and Risk Reduction.” Annual Review of Resource Economics 11(1): 399–418. https://doi.org/10.1146/annurev-resource-100518-094028.
    OpenURL
  38. ↵
    Kousky, Carolyn, and Karina French. 2023. Inclusive Insurance for Climate-Related Disasters: A Roadmap for the United States. January 24. Ceres. https://www.ceres.org/resources/reports/inclusive-insurance-roadmap.
  39. ↵
    Landry, Jeff. 2024. “The Fight for Common Sense in Our Insurance Crisis.” August 1. Office of the Governor. https://gov.louisiana.gov/index.cfm/newsroom/detail/4594.
  40. ↵
    Lee, Jee Young, and Shannon Van Zandt. 2019. “Housing Tenure and Social Vulnerability to Disasters: A Review of the Evidence.” Journal of Planning Literature 34(2): 156–70. https://doi.org/10.1177/0885412218812080.
    OpenURLCrossRef
  41. ↵
    Legal Services Corporation. 2022. The Justice Gap: The Unmet Civil Legal Needs of Low-Income Americans. April. Legal Services Corporation. https://lsc-live.app.box.com/s/xl2v2uraiotbbzrhuwtjlgi0emp3myz1.
  42. ↵
    Li, Yueqi, Lauren Clay, Haley Murphy, Hao-Che Wu, and Alex Greer. 2024. “When Do Households Evacuate? Hurricane Evacuation Timing During the COVID-19 Pandemic.” Environmental Hazards 23(5): 506–526. https://doi.org/10.1080/17477891.2024.2333739.
    OpenURL
  43. ↵
    Lin, Xiao, Mark J. Browne, and Annette Hofmann. 2022. “Race Discrimination in the Adjudication of Claims: Evidence from Earthquake Insurance.” Journal of Risk and Insurance 89(3): 553–80. https://doi.org/10.1111/jori.12386.
    OpenURL
  44. ↵
    Louisiana Department of Insurance. 2021. Commissioner Donelon Updates 2020 Hurricane Data Through Sept. 30, 2021. Louisiana Department of Insurance.
    Google Scholar
  45. ↵
    Louisiana Legislative Auditor. 2022. Residential Property Insurance Complaint Statistics. Louisiana Department of Insurance. https://app.lla.state.la.us/publicreports.nsf/0/a78b15fc0c7658d386258845007cb63c/$file/00026be3a.pdf
  46. ↵
    Marino, Elizabeth K., and Alexander J. Faas. 2020. “Is Vulnerability an Outdated Concept? After Subjects and Spaces.” Annals of Anthropological Practice 44(1): 33–46. https://doi.org/10.1111/napa.12132.
    OpenURL
  47. ↵
    Maschke, Alena. 2024. “Four Years after Hurricane Laura Lake Charles Has Recovered but Many Are Left Behind.” Acadiana Advocate, August 31. https://www.theadvocate.com/acadiana/news/four-years-after-hurricane-laura-lake-charles-has-recovered-but-many-are-left-behind/article_4d591620-6711-11ef-a321-4b1cbfe951d5.html.
  48. ↵
    Miller, Richard E., and Austin Sarat. 1981. “Grievances, Claims, and Disputes: Assessing the Adversary Culture.” Law & Society Review 15(3–4): 525–66. https://doi.org/10.2307/3053502.
    OpenURL
  49. ↵
    Mower, Lawrence. 2023. “Florida Leaders Blame Insurance Crisis on Lawsuits, but Evidence Is Thin.” Tampa Bay Times, October 19. https://www.tampabay.com/news/florida-politics/2023/10/19/florida-leaders-blame-insurance-crisis-lawsuits-evidence-is-thin/.
  50. ↵
    Muñoz, Cristina E., and Eric Tate. 2016. “Unequal Recovery? Federal Resource Distribution After a Midwest Flood Disaster.” International Journal of Environmental Research and Public Health 13(5) 507. https://doi.org/10.3390/ijerph13050507.
    OpenURL
  51. ↵
    Myrick, Amy, Robert L. Nelson, and Laura Beth Nielson. 2012. “Race and Representation: Racial Disparities in Legal Representation for Employment Civil Rights Plaintiffs.” NYU Journal of Legislation and Public Policy 15: 705.
    Google ScholarOpenURL
  52. ↵
    Nalla, Vineetha, Nihal Ranjit, and Garima Jain. 2021. “Representations of Disaster Recovery Needs: A Study of Legal Frameworks and Litigation in Odisha.” International Journal of Disaster Risk Reduction 57: 102163. https://doi.org/10.1016/j.ijdrr.2021.102163.
    OpenURL
  53. ↵
    National Institute of Standards and Technology and Applied Research Associates (NIST/ARA). 2020. NIST/ARA Hurricane Laura Rapid Response Windfield Estimate: Notes for Preliminary Windfield Release 3. September 4. Design Safe. https://www.designsafe-ci.org/data/browser/tapis/designsafe.storage.community/%2FRecon%20Portal%2F2020%20Hurricane%20Laura%20Cameron%20Louisiana%20USA%2FNIST%20ARA%20Rapid%20Reponse%20Windfield%20Estimate.
  54. ↵
    Pasch, Richard, Robbie Berg, David Roberts, and Philippe Papin. 2021. Hurricane Laura (AL132020): National Hurricane Center Tropical Cyclone Report. May 26. National Oceanic and Atmospheric Administration. https://www.nhc.noaa.gov/data/tcr/AL132020_Laura.pdf.
  55. ↵
    Peacock, Walter Gillis, Betty Hearn Morrow, and Hugh Gladwin. 1997. Hurricane Andrew: Ethnicity, Gender, and the Sociology of Disasters. Routledge.
    Google Scholar
  56. ↵
    Peacock, Walter Gillis, Shannon Van Zandt, Yang Zhang, and Wesley E. Highfield. 2014. “Inequities in Long-Term Housing Recovery After Disasters.” Journal of the American Planning Association 80(4): 356–71. https://doi.org/10.1080/01944363.2014.980440.
    OpenURLCrossRef
  57. ↵
    Raker, Ethan J. 2023. “Stratifying Disaster: State Aid, Institutional Processes, and Inequality in American Communities.” Social Forces 102(2): 430–53. https://doi.org/10.1093/sf/soad050.
    OpenURL
  58. ↵
    Rhodes, Anna, and Max Besbris. 2022. Soaking the Middle Class: Suburban Inequality and Recovery from Disaster. Russell Sage Foundation.
    Google Scholar
  59. ↵
    Rivera, Danielle Zoe, Bradleigh Jenkins, and Rebecca Randolph. 2022. “Procedural Vulnerability and Its Effects on Equitable Post-Disaster Recovery in Low-Income Communities.” Journal of the American Planning Association 88(2): 220–31. https://doi.org/10.1080/01944363.2021.1929417.
    OpenURL
  60. ↵
    Sandefur, Rebecca L. 2008. “Access to Civil Justice and Race, Class, and Gender Inequality.” Annual Review of Sociology 34: 339–58.
    Google ScholarOpenURLCrossRef
  61. ↵
    Sandefur, Rebecca L. 2012. “Money Isn’t Everything: Understanding Moderate Income Households’ Use of Lawyers’ Services.” In Middle Income Access to Justice, edited by M. Trebilcock, Anthony Duggan, and Lorne Sossin. University of Toronto Press.
    Google Scholar
  62. ↵
    Sandefur, Rebecca L. 2014. Accessing Justice in the Contemporary USA: Findings from the Community Needs and Services Study. August 8. American Bar Foundation. https://www.americanbarfoundation.org/wp-content/uploads/2023/04/sandefur_accessing_justice_in_the_contemporary_usa._aug._2014.pdf.
  63. ↵
    Sandefur, Rebecca L. 2019. “Access to What?” Daedalus 148(1): 49–55. https://doi.org/10.1162/daed_a_00534.
    OpenURL
  64. ↵
    Schwarcz, Daniel. 2008. “Redesigning Consumer Dispute Resolution: A Case Study of the British and American Approaches to Insurance Claims Conflict.” Tulane Law Review 83(3): 735.
    Google ScholarOpenURL
  65. ↵
    Schwarcz, Daniel. 2019. “Towards a Civil Rights Approach to Insurance Anti-Discrimination Law.” DePaul Law Review 69(2): 657–98.
    Google ScholarOpenURL
  66. ↵
    Silbey, Susan S., Patricia Ewick, Elizabeth Schuster, and Lisa Kaunelis. 1993. Differential Use of Courts by Minority and Non-Minority Populations in New Jersey. November. New Jersey Judiciary, Supreme Court Task Force on Minority Concerns. https://www.ojp.gov/pdffiles1/Digitization/149226NCJRS.pdf.
  67. ↵
    Simington, Jasmine. 2023. “Displaced Trust: Disrupting Legal Estrangement During Disaster Recovery.” Social Forces 102(2): 771–89. https://doi.org/10.1093/sf/soad026.
    OpenURL
  68. ↵
    Squires, Gregory D. 2003. “Racial Profiling, Insurance Style: Insurance Redlining and the Uneven Development of Metropolitan Areas.” Journal of Urban Affairs 25(4): 391–410. https://doi.org/10.1111/1467-9906.t01-1-00168.
    OpenURL
  69. ↵
    Squires, Gregory D., Sally O’Connor, and Josh Silver. 2001. “The Unavailability of Information on Insurance Unavailability: Insurance Redlining and the Absence of Geocoded Disclosure Data.” Housing Policy Debate 12(2): 347–72. https://doi.org/10.1080/10511482.2001.9521409.
    OpenURL
  70. ↵
    Sterett, Susan M., and Laura K. Mateczun. 2020. “Displacement, Legal Mobilization, and Disasters: Trial Courts and Legal Process.” Risk, Hazards & Crisis in Public Policy 11(4): 348–76. https://doi.org/10.1002/rhc3.12199.
    OpenURL
  71. ↵
    Sterett, Susan M., and Laura K. Mateczun. 2022. “Legal Claims and Compensation in Climate-Related Disasters.” Political Science Quarterly 137(2): 293–330. https://doi.org/10.1002/polq.13320.
    OpenURL
  72. ↵
    Tierney, Kathleen. 2012. “Disaster Governance: Social, Political, and Economic Dimensions.” Annual Review of Environment and Resources 37(1): 341–63. https://doi.org/10.1146/annurev-environ-020911-095618.
    OpenURLCrossRefWeb of Science
  73. ↵
    Transactional Records Access Clearinghouse (TRAC). 2021. One Year After Hurricane Laura, Louisiana Plaintiffs Rush to Meet One-Year Insurance Filing Deadline in Federal Court. September 17. Transactional Records Access Clearinghouse. https://tracreports.org/tracreports/civil/661/.
  74. ↵
    Transactional Records Access Clearinghouse (TRAC). 2022. Insurance Lawsuits Skyrocket in Communities Hit Hard by Extreme Weather Events. October 19. Transactional Records Access Clearinghouse. https://tracreports.org/whatsnew/email.221019.html.
  75. ↵
    US Census Bureau. 2024. Methodology for the United States Population Estimates: Vintage 2024. December. US Census Bureau. https://www2.census.gov/programs-surveys/popest/technical-documentation/methodology/2020-2024/methods-statement-v2024.pdf.
  76. ↵
    Veland, Siri, Richard Howitt, Dale Dominey-Howes, Frank Thomalla, and Donna Houston. 2013. “Procedural Vulnerability: Understanding Environmental Change in a Remote Indigenous Community.” Global Environmental Change 23(1): 314–26. https://doi.org/10.1016/j.gloenvcha.2012.10.009.
    OpenURL
  77. ↵
    Western District of Louisiana. 2023. Case Management Order No. 1. Western District of Louisiana. https://www.lawd.uscourts.gov/sites/lawd/files/UPLOADS/Laura%20Delta%20-%20CMO_0.pdf.
  78. ↵
    You, Xuesong, and Carolyn Kousky. 2024. “Improving Household and Community Disaster Recovery: Evidence on the Role of Insurance.” Journal of Risk and Insurance 91(2): 299–338.
    Google ScholarOpenURL
  79. ↵
    Zink, Kathleen A. 2014. “Should Neither Wind nor Rain nor Hurricane Keep Victims from Recovery?: Examining the Tort and Insurance Systems’ Ability to Compensate Hurricane Victims.” Fordham Law Review 83(3): 1621–67.
    Google ScholarOpenURL
Loading
Loading
Loading
Loading
PDF
Back to top

In this issue

RSF: The Russell Sage Foundation Journal of the Social Sciences

RSF: The Russell Sage Foundation Journal of the Social Sciences

Vol. 12, Issue 4

1 Jul 2026

Table of Contents Table of Contents (PDF) Cover (PDF) Index by author Front Matter (PDF)

Jump to section

Keywords

insurance, hurricane damage, civil litigation, claims disputes, procedural vulnerability, Southwest Louisiana, sociodemographic difference