Site sourcing and screening — Arkansas

What the screen looks like before you can even file

There is no separately named diligence phase in an Arkansas deal any more than there is in California's or Texas's. Reading the Arkansas Development Finance Authority's (ADFA) own Qualified Allocation Plan and its attached Multifamily Housing Application Guidelines makes clear how much of this phase's "screening" is threshold-requirement compliance work wearing a different hat. Over roughly two to six weeks, a developer or analyst works through the following before the Multifamily Housing Application (MFHA) can even be assembled.

Roughly half of that list is arithmetic and lookups; the other half is paperwork with a clock attached to it.

As in California and Texas, two decisions come out of this phase: whether to pursue the site, and what goes into the option or contract. Arkansas's own twist on the second decision is date-shaped rather than price-shaped. The site control instrument has to run, without extension fees, to a fixed calendar date — not to expire prior to December 6 of the year of the MFHA (Application Guidelines § I(C)(4)) — tied to the following year's carryover cycle rather than a rolling term counted from signing.

One more timing fact worth building into an Arkansas calendar. This is a single-regulator program — ADFA alone, not California's three-way CTCAC/CDLAC/HCD split — but the QAP still moves on its own annual clock, and the clock is not always finished when the document is posted. The version on ADFA's site as of this research, the "2027 QAP" uploaded in September 2026, carries a blank Board-adoption signature block; the prior year's file is both named and signed as executed. Until ADFA republishes a signed version, its point values and thresholds are best treated as the working draft ADFA staff are using, not a locked document.

Zoning is permissive twice over, and the letter assumes it exists

Arkansas's zoning story splits differently from both California's — every one of 539 jurisdictions zones, vintage is the problem — and Texas's, where counties categorically cannot zone at all. In Arkansas, both municipal and county zoning are optional powers a governing body may choose to exercise, not a duty imposed on it. A municipality's governing body may adopt a zoning ordinance under Ark. Code Ann. § 14-56-416; nothing in that section compels it to. The same permissive structure runs one level up: under §§ 14-17-203 and 14-17-209, a county judge may appoint a county planning board, which may in turn prepare and adopt a zoning ordinance for all or part of the county's unincorporated area — but until a county actually does that, its unincorporated land carries no zoning classification to look up, by design, not by data gap.

A parcel a half mile outside a mid-size city's limits can sit inside that city's planning and zoning reach even though it reads as "unincorporated" on a plat. The reverse also holds: a parcel outside every municipality's reach, in a county that has never created a planning board, is genuinely unzoned. The QAP's own zoning-letter requirement doesn't spell out what satisfies it in that case. Unlike Texas's QAP, which names a specific "No Zoning Ordinance in Effect" letter type for exactly this scenario, Arkansas's Application Guidelines ask only for "a signed letter... from the appropriate zoning authority... stating... the property is properly zoned for such proposed use" (§ I(C)(5)) — text that assumes an authority and an affirmative zoning designation both exist to certify.

Arkansas has 75 counties and roughly 500 incorporated cities and towns as of the 2020 Census, each independently deciding whether to exercise the zoning power at all — a different failure mode from California's amendment-vintage problem, but the same practical result for a screening tool: the answer to "what does the code say" is sometimes "there is no code."

The parcel layer is one live service, with real seams

Arkansas's county assessors operate under the statewide oversight of the Arkansas Assessment Coordination Division, a division of the Department of Finance and Administration that supervises assessors and equalization boards and runs a statewide reappraisal-cycle and ratio-study program. What that buys a screening tool is closer to California's aspiration of one statewide layer than California itself has achieved: a single, live, statewide parcel service, published by the Arkansas GIS Office and built from the counties' own Computer Aided Mass Appraisal systems through the County Assessor Mapping Program (CAMP).

Vintage tells the same story California's Riverside zoning layer tells, on a different layer. The whole service's own "last updated" metadata reads a single date, April 1, 2026 — but the underlying camadate field, each county's own CAMA extraction date, actually ranges from September 24, 2025 to March 17, 2026. A roughly six-month spread of real county-level currency sits hidden behind one republish stamp.

Floodplain is an absolute bar, and the environmental review comes before the filing, not after

Arkansas does not treat floodplain risk as a scoring factor or a mitigation queue the way California's SB 35 exclusion list or Texas's Undesirable Site Feature buffers do. It is a flat threshold requirement, stated without qualification: "No federally designated 100-year floodplain or floodway sites allowed. ADFA will not fund any project on a parcel that contains a 100 year floodplain or floodway... No exceptions" (Application Guidelines § I(C)(4)). An existing, operating affordable housing development already inside a floodplain can still apply for acquisition or rehabilitation, but only if it has not flooded since construction and carries flood insurance for the full affordability period — new construction has no comparable path. HOME- and NHTF-funded developments layer a second federal standard on top, the Federal Flood Risk Management Standard.

ADFA's own January 26, 2024 memorandum adds a timing requirement with no direct California or Texas analog: the environmental review itself — not merely an order for one — has to be complete, performed by a qualified environmental professional, before the MFHA is submitted. New construction of more than four units, or rehabilitation that doesn't meet the narrow categorical-exclusion criteria at 24 CFR § 58.35(a), needs a full Environmental Assessment; smaller or more limited rehabilitation needs the lighter Categorically-Excluded-Subject-to-§58.5 (CEST) review instead. Either way, the review has to be finished pre-submission, and the requirement applies to every Tax Credit applicant, not only ones layering in HOME or National Housing Trust Fund money.

Absence of a hit on the layers actually queried here says nothing about the ones this research didn't have time to test — the same caution California's own screen draws about DTSC and SWRCB.

Site Selection is 27 points of measured distances, and Location is another 10

Unlike California's Opportunity Area designation or Texas's Neighborhood Risk Factor gates, Arkansas folds location quality into one self-contained scoring item. "Location" is worth up to 10 points, computed by ADFA itself for every census tract from three inputs — unemployment rate, vacancy rate, and population growth — published on ADFA's own ArcGIS Experience app and re-versioned annually (the current QAP calls this release "AOI 2027"). ADFA does the calculation; a developer only has to look the number up (Application Guidelines § II(A), Item 1).

A Qualified Census Tract location is worth exactly one Arkansas scoring point, and only if it's paired with a documented community revitalization plan that specifically addresses affordable housing (Application Guidelines § II(A), Item 12) — a far smaller in-QAP reward than California's opportunity-area mechanics. The number that actually matters for a QCT site sits outside ADFA's own point table entirely: the federal 130% eligible-basis increase under 26 U.S.C. § 42(d)(5)(B), which applies automatically regardless of anything ADFA scores. A QCT location also moves a project up Arkansas's own State LIHTC tie-breaker priority order, behind counties in the state's job-creation incentive Tiers 3 or 4 and ahead of counties without a recent federal LIHTC award (2027 QAP § IV).

None of this location math matters if the application can't clear ADFA's own floor: a minimum self-score of 75 points to apply for 9% credits, or 45 points for Bond Volume Cap/4% credits, stated flatly as "No exceptions" (Application Guidelines § I(C)(31)). The QAP's own scoring table leaves its "Total Points Possible" cell blank; summing each item's own stated maximum gives 99, but that is this guide's arithmetic on the published numbers, not an ADFA-stated ceiling — several point categories are mutually exclusive alternatives no single project could actually stack together.

Where the screen stops, and what it cannot tell you

None of these figures is a construction-cost estimate any more than California's Threshold Basis Limits are; they cap what ADFA will fund or reward, and say nothing about what a general contractor will actually quote for a given site.

ADFA's own award history is a real gap next to California's single combined List_of_Projects.xlsx or even Texas's per-round XLSX file. Each year since 2005 gets its own PDF roster — the 2026 9% round shows 12 developments awarded — with applicant, developer, county, congressional district, unit counts, and federal/HOME/NHTF dollar amounts, but no parcel identifier, no census tract field, and no spreadsheet to query across years. ADFA describes its annual 9% program as roughly $9 million in federal credit authority; the QAP itself caps any single 9% development at $1,300,000 of that annual ceiling (2027 QAP § III(C)).

One genuine simplification relative to California: ADFA does not publish its own income-and-rent-limit PDFs the way CTCAC does. Applicants read HUD's own Multifamily Tax Subsidy Project (MTSP) limits directly — a screening tool built for Arkansas doesn't need to solve CTCAC's three-vintage PDF problem at all, because there is no ADFA-specific layer standing between the site and HUD's own published numbers.

The honest limit is the same one California's and Texas's own guides draw. Whether the seller actually holds the option to December 6, whether a county that has never adopted a zoning ordinance will issue a letter ADFA will accept, whether a competing application is filing in the same round — none of that is a dataset, and none of it should be synthesized. What screening can and should do in Arkansas is make sure the arithmetic questions are answered before the phone call: floodplain status, the Site Selection and Location point math against the 75-point floor, and whether the environmental review can realistically finish before the deadline. EZFeasi has no Arkansas parcel, zoning, hazard, or award data loaded today — this entire phase is greenfield product work, not a refinement of an existing Arkansas feature.