The Lease Arrives First: Mapping Work-Authorization Lag and Rent Risk Across US Refugee Metros

In refugee resettlement, the ninety-day clock is usually described as a service window. It is also a rent clock. A family may sign a lease, owe a security deposit, and face the first month’s payment before the employment card that many landlords and employers treat as proof of work eligibility has arrived. The gap is not a single national number; it varies by geography, caseload, and the administrative machinery that assigns paperwork to offices. Mapping that gap is a modest but useful exercise for advocates — provided the map remains an aggregate warning surface rather than a ledger of individual arrivals.

Why the ninety-day window is also a housing window

Refugees admitted to the United States are employment authorized incident to their status. In practice, however, many employers and property managers ask for an Employment Authorization Document, and many refugees apply for one to satisfy I-9 verification and landlord screening. The Reception and Placement period gives resettlement agencies roughly ninety days to help with housing, employment, and orientation. During that same period, leases begin, utility deposits come due, and caseworkers must reconcile a family’s budget against an income stream that has not yet started.

The policy question is therefore not whether a refugee may work. It is when the documentary evidence of that right reaches the household, and how that timing interacts with local rent levels. A metro where published processing times are long and rents are high against modest entry-level wages produces a wider risk window than a metro where one of those conditions is absent. The map should show that interaction, not adjudicate individual cases.

What public data can actually support

A defensible national audit can be built from four public layers. First, USCIS publishes case processing times for Form I-765 by category and office, typically as ranges. These are not individual wait times, but they can be converted into a metro-level lag estimate by assigning each resettlement metro to the field office or service center that handles its cases, then using the midpoint and upper bound as sensitivity checks. Second, the Refugee Processing Center publishes aggregate arrivals by state and, in some tables, by metropolitan area. Third, HUD Fair Market Rents and income limits provide a consistent rent-burden threshold; the common benchmark is rent exceeding thirty percent of household income. Fourth, the Office of Refugee Resettlement and local resettlement agencies report caseload counts, though not always at the same geographic scale.

The weaknesses are real and should be printed alongside the map. Processing-time ranges are administrative estimates, not distributions. Not every refugee applies for an EAD, and some arrive with documents that employers accept without one. Field-office assignments can shift, and a metro may straddle more than one jurisdiction. Rent-burden calculations require an income assumption, which is especially unstable in the first ninety days. These are reasons for explicit uncertainty bands, not reasons to abandon the exercise.

A three-layer map, not a dragnet

The most useful design is a three-layer composite at the metro or county level. The first layer is authorization lag: the estimated number of days between arrival and documented work eligibility, drawn from published processing ranges. The second is rent pressure: median gross rent as a share of an entry-level wage, or Fair Market Rent against a conservative household-income estimate. The third is caseload density: arrivals per resettlement agency or per available caseworker, which indicates how much strain the local service system faces during the same ninety days.

Those layers can be combined into a simple lease-gap index, but the index should be presented as a set of bins — low, moderate, elevated, high — rather than a precise score. Small cells should be suppressed. No layer should use client addresses, agency case files, or individual immigration receipt numbers. The unit of analysis is the metro, not the household. That distinction is not merely ethical decoration; it is what keeps the product from becoming a surveillance aid for landlords or enforcement agencies.

Reading the map without overreading it

An aggregate map cannot establish that a particular family will miss rent. It can show where the structural conditions for a housing crunch are concentrated. A metro with a long published EAD range, a tight rental market, and a high arrival-to-caseworker ratio deserves attention even if many individual families manage without difficulty. Conversely, a metro with a short processing range and lower rents may still have severe housing problems rooted in credit history, landlord discrimination, or informal rental arrangements that no public dataset captures.

Advocates should use the map to ask better questions rather than to make forecasts. Where are legal-aid providers already fielding rent-deadline calls? Are local USCIS field offices issuing EADs at the lower or upper end of their published range? Do property managers accept alternative documentation, such as an I-94 showing refugee status, in place of a card? The map can frame those questions; it cannot answer them alone.

What advocates can do with the layer

A practical use is budget justification. Resettlement agencies and legal-aid offices can pair the lease-gap index with local case counts to argue for short-term rental assistance, landlord mediation, or expedited document clinics. Another use is monitoring: if published processing times change in 2026, the map can be updated quarterly to show whether the risk window is widening or narrowing. A third is public education. Landlords and employers often do not know that refugees are work authorized at admission; a map that explains the documentary bottleneck can support local guidance without exposing any individual’s status.

The protocol should be reproducible. Publish the join keys, the date of each data pull, the assumptions behind the income estimate, and the suppression rules. Version the map. Keep the underlying code open. Do not add a searchable address field. The goal is a map that helps a city see where the ninety-day clock and the rent clock collide, while leaving the households behind those numbers where they belong: outside the frame.