Economic and Housing Recovery Three Years After the Maui Wildfires: Findings from the Maui Recovery Survey

Executive Summary

Three years after the August 2023 wildfires on Maui, substantial steps forward have been made, but important gaps remain. Housing has stabilized for many affected families, with more households in permanent homes and some able to return to their communities. But employment and incomes have moved little over the past year. This report presents findings from the Maui Recovery Survey, an ongoing survey of people who lived, worked, or owned a business in West Maui or Kula at the time of the disaster.

Displacement has declined modestly. The share of households no longer displaced rose modestly over the past year, and fewer are displaced from both their home and West Maui. Of those displaced from both their home and West Maui in early 2025, about one in six have returned to West Maui but not yet to their home. Displacement also continues to extend beyond the burn area, where the share displaced from their home but remaining in the same area has risen.

Housing stability has continued to improve. Homeownership has recovered slightly from 38% to 42%. For some, relief programs met needs that predated the fires. About 60% of those who were unhoused or in informal arrangements before the disaster had a rental by 2025, and a further 20% by 2026. About two-thirds of fire-affected households now live in permanent housing, up from just over half in early 2025. Households are also moving far less than in the first two years, and among those still in temporary arrangements the share at the same address for more than a year has tripled to almost two-thirds. However, overall rates of temporary and unstable housing remain elevated.

Rental costs have eased for smaller properties, but remain heightened for larger units. Median rent for studios and one-bedrooms has fallen to roughly its pre-fire level by May 2026. Rents for larger units have stayed broadly flat and well above pre-disaster levels. Overall, almost half of fire-affected renters now pay less than before the fires (largely because of rental assistance), but at the same time almost 40% pay more. The share of households receiving full or near-full rental assistance has more than halved, from 37% in January 2025 to 15% in May 2026.

Employment continues to be an area of limited progress. 42% of fire-affected people work full-time, unchanged from a year earlier and still well below the 62% who did so before the fires. Unemployment has fallen from 10% to 8%, but exits from the labor market through retirement or no longer seeking work have risen from 26% to 30%. Even among working-age people, more than one and a half times as many are out of the labor force as before the disaster. Among people employed in the visitor industry before the fires, about two-thirds still work in the sector, with little change over the last year.

Poverty levels have risen since the wildfires, but some improvements are emerging. The share of households in poverty has declined modestly over the past year but remains markedly above pre-fire levels, and most households are still poorer than before. The share of households above but near the poverty threshold has increased the most after the wildfires, but it has also been on a downward trajectory in the last few months.

Unmet needs have declined slightly, but assistance has fallen substantially. Financial assistance remains the most frequently reported need. FEMA assistance halved over the last year and a half. Other government assistance has also declined, and only half as many people now receive help from community organizations compared with the start of 2025. About half of fire-affected people received no assistance of any kind in May 2026, almost double the share a year earlier. Some of this decline may reflect households regaining stability, but the scale suggests assistance has likely also fallen for households who still need it.

With support from the Hawai‘i Community Foundation, UHERO will continue to monitor and analyze the progress of the recovery in its fourth year.

Introduction

On August 8, 2023, wildfires devastated Lahaina and parts of Kula. They claimed many lives, destroyed homes, public infrastructure, cultural resources, businesses, jobs and livelihoods and displaced thousands of residents. UHERO, with support from the Hawai‘i Community Foundation, has tracked the disaster recovery with the Maui Recovery Survey (MRS). The MRS is an ongoing survey of individuals who lived, worked, or owned a business in West Maui or Kula at the time of the disaster. Reliable and timely data on how fire-affected people are faring is still limited. Official statistics from the Census Bureau for Lahaina that distinguish between the time before and after the wildfires will not be available until 2030 and they are not designed to isolate the experiences of the disaster-affected population. The MRS fills this gap for the wildfire-affected community. The survey was run on a monthly cadence between August 2024 and July 2025 and bimonthly since August 2025. Participants can join at any point and are not required to respond every data collection period.

The survey is designed to track change over time and to reflect the lived experience of recovery. Data is collected on housing status, employment, income, unmet needs and assistance. Almost a thousand people have taken part so far. This report draws on time series data from January 2025 to May 2026 from 563 respondents who have provided at least one response in each year that the data collection has run. In addition, a longitudinal dataset has been created that follows 465 survey participants over time and compares their responses from the first half year of 2025 to the first half of 2026. A detailed summary of the survey and methodology is provided in the appendix, along with an overview of sample demographics.

Displacement

Displacement has declined for West Maui residents over the past year. For people who lived in West Maui at the time of the fires, more households are not or no longer displaced (up from 23% to 27%, Figure 1). Fewer families are displaced from both their home and from West Maui (44% in 2025 compared to 37% in 2026), while the share who reside in West Maui without returning to their pre-fire home has risen from 33% to 37%.

Of those who were displaced from their homes and West Maui in the first half of 2025, about 1 in 6 have been able to return to West Maui, but not (yet) to their home (Figure 2). Only about 7% of formerly displaced households have been able to return to their homes in West Maui. Of those who returned, about half were already residing in the area. While these changes point to modest improvements, displacement remains widespread among fire-affected households.

Displacement also continues to extend beyond the immediate burn area. Among fire-affected people who lived elsewhere on Maui but worked or owned businesses in West Maui or Kula, 12% remain displaced from their homes and area, compared to 15% a year earlier (Figure 1). But at the same time, the share of households displaced from their home but remaining in the same area has risen from 13% to 19%. This pattern is consistent with continued displacement driven by affordability and availability pressures.

These estimates are likely to understate the full extent of displacement. Survey outreach operates primarily through Maui-based networks, and people who have left the islands entirely are likely underrepresented among respondents.

Housing stability

Housing stability has continued to improve, but renting and temporary arrangements remain common. Of fire-affected households who owned their homes before the wildfires, almost a third were still in rentals or lost stable housing altogether by 2025 (Figure 3). The most common transition was from owning to renting. By 2026, the homeownership rate increased from 38% in 2025 to 42% and only very few homeowners had to move out of their homes over the last year. The fires also triggered a substantial increase in unstable housing where families moved in with family or friends or became unhoused. At the same time, for some, disaster relief programs provided essential assistance that was needed long before the fires. Of people who were unhoused or in informal arrangements with family or friends prior to the disaster, about 60% were housed in a rental by 2025. Of people who were still unhoused in 2025 but got access to a rental over the last year, the majority were former owners or renters who regained stable housing, but almost 20% were people who had been unhoused since before the fires.

In addition, the distribution of housing permanency has shifted markedly over the past year and a half (Figure 4). Temporary housing, which accounted for 37% of fire-affected households in January 2025, has declined steadily and now accounts for less than 30%. Over the same period the share living in a permanent pre-fire home has risen from 36% to 45%. The share of households living in a permanent new home has been broadly flat over the period. This reflects the pace of reconstruction, with 315 homes remaining under construction in the fire-affected areas as of July 2026 (County of Maui, 2026). Overall, almost two-thirds now live in permanent housing, whether a pre-fire home or a new one, up from 56% in early 2025.

Households are also moving far less than they did in the first two years after the fires, when many cycled through multiple temporary arrangements. Among those in temporary arrangements, the share who have lived at their current address for more than a year has risen from about one in five in January 2025 to almost two in three in May 2026 (Figure 5). Repeated moves between short-term placements, which characterized much of the first two years after the fires, have become far less common, and some households have had some level of continuity to re-establish routines and community belonging.

Rent

Three years after the fires, there have been some encouraging signs of easing rental costs especially for smaller properties, but at the same time changes in access to rental assistance have increased pressures.

Since the beginning of 2025, the share of fire-affected households receiving full rental assistance (defined as paying less than $100 per month in rent) has halved. In January 2025, 37% of households received full rental assistance, but by May 2026 this was down to 15%.

The decline in full rental assistance has affected every unit size. Among households in studios and one-bedroom units, the share paying zero or near-zero rent has fallen from almost half of renters in January 2025 to below a quarter in May 2026. For larger unit sizes, rates were lower to begin with, but continued to decline (from about 30% to around 10% for both two-bedroom and three or more-bedroom units). While some of these trends may be associated with the winding down of support programs and reductions in assistance levels, much of it also likely reflects families moving out of temporary housing that provides full rental assistance, because they need to regain more stability in a permanent home to move on with their lives. Consequently, the gap between what the market charges and what households pay has narrowed.

It is likely that further reductions in access to rental assistance will follow. From September 1, 2026, FEMA will charge Direct Housing occupants 50% of the HUD Fair Market Rent for Maui County (FEMA, 2026), replacing the minimal contributions most households currently pay. Federal assistance is then due to end in February 2027, while the state’s Ka La‘i Ola program is scheduled to run until August 2029 (Office of the Governor, 2026). The survey will continue to monitor the implications of these changes.

Because fewer households receive full rental assistance, typical out-of-pocket rent payments have risen across every unit size (Figure 7). But for smaller units rental cost pressures have actually eased. Households in fully subsidized studios or one-bedroom units continue to pay substantially below the typical pre-fire rental cost for such units (about $700 now compared to about $1,200 before the fires). Households that do not receive full rental assistance pay close to what they would have paid before the fires, as median rent for small units has fallen steadily, from about $2,000 in January 2025 to about $1,200 in May 2026.

However, larger units have not followed in the same way. Median rent for unassisted renters in two-bedroom units still sits at above $2,300, more than 45% above the pre-fire median rent of $1,600. For three-bedroom units median rent for households without full rental assistance has been broadly flat at about $2,500 to $2,700, well above the pre-fire figure of $1,400. Families needing more space therefore face a market that has not eased, with rents for larger units still well above pre-fire levels. As expected, the distribution of unit sizes has not changed much at the population level, suggesting this is not a decisive factor. While in 2026 more households have moved into smaller units than they had before the fires, at the same time a higher share of households are now in larger properties than before the disaster, suggesting limited changes overall (Table 1).

Table 1: Change in unit sizes

The data includes households who responded at least once both in the first half of 2025 and in the first half of 2026. Because the same households are compared at both points, differences reflect changes in their circumstances rather than changes in who responded.

Overall among all fire-affected renters, 46% pay less than they did before the fires and 39% pay more (Figure 8). Among renters without full assistance, the proportions reverse: 31% pay less and 50% pay more rent than before the disaster. Notably, the share of households who pay more in rent has declined somewhat over the last year, with a corresponding rise in households who pay less.

The rental market has moved in favor of fire-affected households at the smaller end, where prices have returned to pre-fire levels, but little has changed for households needing larger properties. At the same time, under current conditions, the assistance that has kept out-of-pocket costs low is being withdrawn faster than market rents are falling.

Employment

Among fire-affected people, 42% work full-time, unchanged from a year earlier but still well below the 62% who did so before the fires. The pattern holds among those under 65, where 51% currently work full-time against 74% before the fires.

Two employment measures have improved. The share of people who are not employed and seeking work has fallen from 10% to 8% over the past year, and from 12% to 10% among people under 65. Unemployment among fire-affected people nonetheless remains well above the county-wide rate of 2.5% (UHERO Data Portal, 2026). The share of fire-affected people working part-time but wanting more hours has fallen slightly from 14% to 13%, and from 16% to 15% among those under 65, but remains more than three times the pre-fire level of 4%.

However, at the same time, exits from the labor market due to retirement or no longer seeking work continue to increase, from 26% to 30% over the past year, against 23% before the fires. Notably, among people under 65 the same measure has risen from 14% to 16%, compared with 10% before the fires. Withdrawal from the labor market is therefore not confined to those reaching retirement age: among working-age fire-affected people, more than one and a half times as many are out of the labor force as before the disaster.

Maui’s dominant tourism economy continues to show signs of long-term disruption. Real visitor spending remains nearly 20% below pre-fire levels in the first quarter of the year (UHERO Data Portal, 2026). Among those employed in the visitor industry before the fires, 65% still work in it, down slightly from 67% a year earlier (Figure 10) and fewer people work in other industries. But at the same time, the share of former visitor-industry workers who are not employed has also fallen from 26% to 22%. This suggests that fewer people working in the visitor industries hold additional jobs in other industries. This could either be a sign of lower need for multiple jobs, or given the ongoing income squeeze, it may be associated with lower ability to find and/or hold multiple jobs. Overall, three years on, the sector has not returned to a scale that would re-absorb the workers who had to leave it.

Employment continues to be an area of limited progress, as full-time work is no higher than a year ago, and the improvement in unemployment has come alongside continued movement out of the labor force.

Poverty

These changes in employment have had consequences for people’s incomes. Poverty levels have risen since the wildfires, but recently some improvements are emerging. Poverty is measured by the income-to-poverty ratio. It compares a household’s income to the federal poverty threshold for its size. A ratio below 1 means the household is in poverty, and a ratio between 1 and 2 can be considered near poverty. The share of households in poverty has declined modestly in the last few months compared to a year earlier, but remains markedly above pre-fire levels which was around 11%. The near-poverty group increased most dramatically after the wildfires to rates almost three times higher than before the disaster by January 2025, but it has been on a downward trajectory compared to a year ago. Because of reporting differences between respondents, the poverty levels are best read as indicative and interpretations should focus on changes over time.

Figure 12 compares each household’s poverty status with its own position before the fires. Fewer households were poorer than before the fires in 2026 compared with a year earlier. Very few of those who had already recovered in 2025 slipped back. The better-off group has grown, taking in households from both of the other groups as their incomes recovered. However, the majority of households are still poorer than they were before the fires.

Needs and assistance

Over the last year, unmet needs have fallen, but remain substantial. Financial assistance continues to be the most frequently reported need, affecting about 29% of fire-affected people in 2026, down from 32% a year earlier. Unmet needs for housing and food have only fallen slightly, and the share of people needing help navigating available resources remains largely unchanged at around 12%.

Assistance has fallen substantially. FEMA assistance dropped by half, from about a third of the fire-affected people in January 2025 to 16% in May 2026. More than half of the fire-affected community received FEMA assistance at some point since the wildfires. Receipt of SNAP, WIC or TANF has decreased from 18% to 13% and support from other government sources is down from 9% to 4% over the last year and half. Many community organizations have had fewer resources to provide assistance, as philanthropic funds raised after the fires are drawn down (Hawai‘i Community Foundation, 2026). This has likely contributed to only half as many people receiving assistance from community organizations in May 2026 compared to January 2025. Help from family, friends or individual donors has also fallen from 28% to 16%. It is unsurprising that these informal sources of assistance are not sustainable at scale in the longer term, especially given the economy-wide effects of the disaster.

About half of fire-affected people received no assistance of any kind in May 2026. That is almost double the share of unassisted households in January 2025. Only about 8% report never having received any assistance since the disaster. This illustrates the scale of the reduction in assistance. Part of this decline may reflect households regaining stability, but the magnitude of difference suggests that assistance has likely also fallen for households who still need it.

Conclusion

Three years after the wildfires, the recovery has moved unevenly. Progress in housing stability has emerged, with about two-thirds of fire-affected households now in permanent homes and far fewer families moving between temporary arrangements. But at the same time, employment and incomes have changed little over the last year, and most households are still earning less than they did before the fires. With Hawai‘i Community Foundation’s support, UHERO will continue to monitor the progress of the recovery in its fourth year with the Maui Recovery Survey and other analyses.

Appendix

Methodology

The Maui Recovery Survey (MRS) collects data from people who lived, worked or owned a business in West Maui or Kula during the Maui wildfires in August 2023. Participant recruitment has been facilitated by community partners. Participants receive a $20 gift card for the first survey and a $10 gift card for each monthly follow-up survey that they take. It is not a closed cohort, i.e. eligible people can join anytime. The survey form is available online in six languages: English, Spanish, Tagalog, Ilocano, Tongan, and Vietnamese. Recruitment primarily has relied on multilingual email outreach but is supplemented by physical flyers available in community spaces and at local events. The Hawaiian Council (formerly Council for Native Hawaiian Advancement) provided critical support for community outreach and initial project planning.

The survey questionnaire is designed to follow key indicators such as income, employment, housing type, rent, and residential and job location over time. Participants are asked about their situation before the wildfires, followed by updates on the same set of questions starting in August 2024. For the first year of data collection, the survey used a monthly cadence, switching to bimonthly in August 2025. For analysis, monthly responses from the first year have been organized into two-month periods, so that every data point covers the same length of time. We have used a wide range of data verification checks and individual participant follow-ups to eliminate fraudulent responses from the data. We are confident that this data represents real fire-impacted people. So far almost a thousand people have participated in the survey. This report draws on time series data from January 2025 to May 2026 from 563 respondents who have provided at least one response in each year that the data collection has run. In addition, a longitudinal dataset has been created that follows 465 survey participants over time and compares their responses from the first half year of 2025 to the first half year of 2026.

The survey data has been weighted to maximize representativeness relative to the fire-affected population as a whole. The person-level data has been weighted by income-to-poverty ratio, employment status, tenure and age group, all measured before the wildfires. Iterative proportional fitting (raking) has been used to implement survey weighting. This method continues to adjust the weights until the sample distribution approximately matches the population distributions for the selected weighting variables. Reference data has been obtained from the American Community Survey 5-year sample for 2022 to avoid bias from 2023 data that could have been collected after the fires. Because most of the fire-affected people lived in Lahaina prior to the wildfires, we used data for the Lahaina Census Designated Place as the reference population. Because of substantial sample churn over time, weights have been calculated separately for each two-month period for the time series data, so that each data point is adjusted for the particular mix of people who responded in that period. The average design effect is 1.45 with a margin-of-error of 5.52. Separate person weights have been calculated for the longitudinal dataset used to compare 2025 and 2026 (with a design effect of 1.46 and a margin of error of 5.49) and a pooled version that includes all the respondents in the 2025 and 2026 comparison periods (with an average design effect of 1.45 and margin-of-error of 5.23).

By design, the weighted survey sample closely aligns with the reference population for the weighting variables age, employment, poverty and tenure. Marginally fewer survey respondents were born in Hawai‘i. The survey includes more women than men and due to lack of reference data, we have been unable to compare non-binary people. But given the weighting on key economic metrics, it is unlikely that these differences meaningfully distort any of the findings. 

The analysis relies on weighted descriptive statistics. To visualize time series trends, the raw data has been smoothed using LOESS (Locally Estimated Scatterplot Smoothing), a non-parametric method that fits a line through the data by averaging local subsets of points. This approach does not assume a specific functional form and is useful for identifying gradual changes or inflection points in the data by reducing short-term fluctuations that may be driven by sample churn.

Pooled data results

In the main body of the report, some figures and tables were generated using the longitudinal dataset. Below, similar figures and tables are presented using a pooled dataset, which includes all participants who responded  at least once in the first half of 2025 or the first half of 2026.  This gives a larger overall sample size than the longitudinal data, but because the 2025 and 2026 comparison samples are not identical, differences may partly reflect changes in who responded.

Table 2: Pooled data version of Table 1

Demographics of the survey cohort

Table 3: Baseline sample characteristics before the wildfires

Person-based weighted sample proportions for participants in each dataset (unweighted in brackets) compared with the pre-fire population based on the American Community Survey (ACS) five-year sample for 2018–2022 for the Lahaina Census Designated Place. “In poverty” refers to below the federal poverty threshold, “near poverty” between one and two times the federal poverty threshold and “not in poverty” above two times the federal poverty threshold.

Bibliography

County of Maui. (2026). Rebuild Dashboard [Dataset]. Retrieved July 31, 2026, from https://www.mauirecovers.org/recoverydashboard

FEMA. (2026). Rent increase for Maui wildfire survivors in FEMA temporary housing [Fact sheet]. https://www.fema.gov/fact-sheet/rent-increase-maui-wildfire-survivors-fema-temporary-housing

Hawai‘i Community Foundation. (2026). Maui Strong Fund. https://www.hawaiicommunityfoundation.org/strengthening/maui-strong-fund

Office of the Governor, State of Hawai‘i. (2026, January 23). Gov. Green secures FEMA housing extension for Maui wildfire survivors through February 2027 [Press release]. https://governor.hawaii.gov/newsroom/office-of-the-governor-news-release-gov-green-secures-fema-housing-extension-for-maui-wildfire-survivors-through-february-2027/

UHERO Data Portal. (2026). Total Real Visitor Expenditures [Dataset]. https://data.uhero.hawaii.edu/#/series?id=164727&data_list_id=41&sa=true&geo=MAU&freq=Q