Site sourcing and screening — Alabama

What screening actually looks like in Alabama

Alabama runs its Housing Credit program out of a single agency. The Alabama Housing Finance Authority (AHFA) has been the state's designated administrator of the Housing Credit program since 1987, and unlike California's split between CTCAC (9% scoring) and CDLAC (bond volume cap), AHFA both scores the competitive 9% round and issues the Multifamily Housing Revenue Bonds that carry the 4% credit. One agency, one QAP, one point-scoring addendum. That structure shows up immediately in screening, because the same document that tells you how a site scores also tells you, bluntly, that AHFA intends to fund generally only one project per county each cycle.

The county record and the flood plain check are the closest things to a fast lookup on that list. Everything else is either a phone call to a city that may or may not have a zoning code at all, or a formal third-party deliverable — a Phase I ESA, a market study, a certified boundary survey — that has to be substantially finished, not just directionally sourced, before an application can even be filed. AHFA does not treat these as later-phase paperwork; they are threshold items, reviewed for completeness before the point-scoring evaluation ever starts.

The county-allocation fact deserves its own sentence because it changes what 'screening' even means here. AHFA's stated allocation methodology funds the highest-scoring CHDO-owned project per county toward the 15% CHDO set-aside first, then generally allocates to one project per county until the year's Housing Credits and HOME Funds run out. A site can be objectively excellent and still be structurally unfundable this cycle if a stronger application, or a still-active prior award, has already claimed that county's slot — a constraint no amount of site diligence changes.

The zoning-letter trap is just as concrete. AHFA does not consider a property zoned — for threshold purposes — if final zoning is contingent on further city meetings, approvals, or advertisement, and it requires the evidence in the form of a signed statement from the local jurisdiction. A city planner's verbal assurance that approval is imminent does not satisfy the requirement; the letter has to already be final when the application goes in.

The parcel layer is 67 county offices, not a dataset

There is no unified statewide Alabama parcel database with attributes, and no state agency is charged with building one. Property is assessed annually by each county's own Revenue Commissioner — or, in some counties, a separately elected Tax Assessor — under the general oversight of the Alabama Department of Revenue, which maintains only a directory of links to each county's own site rather than a merged dataset. Most counties run their own public GIS parcel viewer, searchable by owner name, situs address, or parcel number, but each is a separate implementation with its own vendor, its own export format, and its own uptime.

This is structurally the same fleet-of-adapters problem California's SCAG/SANDAG patchwork and Texas's 254 appraisal districts present — a new county is a new integration, and it does not get easier with expansion. What's honest to say here, and worth stating plainly rather than implying otherwise: EZFeasi has not yet built or load-tested adapters against any Alabama county's GIS service, so the specific breakage catalog California's guide can point to by name — a maxRecordCount cap here, a bot-detection shim there — does not yet exist for Alabama. That gap is real, and it is exactly the kind of thing that shows up quietly once ingestion starts.

Zoning is a municipal-only power, and most of the state has none

Alabama's general zoning-enabling statute, Title 11, Chapter 52, Article 4 of the Code of Alabama, gives the power to zone to municipalities — it is a granted power, not a mandatory one, and it says nothing about counties. Alabama counties, as a rule, have no general authority to zone unincorporated land at all. Mobile County is the standing, citable example: its unincorporated territory carries no zoning. Practically, that means Alabama's zoning question is closer to Texas's than California's — the first thing to determine about a site is not what the zone code says, but whether a zone code applies to this parcel in the first place, and for land outside any municipality's limits, the honest answer is very often that it does not.

Alabama's version of extraterritorial reach runs through a municipality's police jurisdiction rather than a dedicated zoning-ETJ statute. Under Ala. Code § 11-40-10, a city's police jurisdiction — the zone in which its ordinances have force outside its own limits — extends 3 miles beyond the corporate limits for cities of 6,000 or more inhabitants, and 1.5 miles for smaller cities and towns. Act 2021-297 froze further expansion of any municipality's police jurisdiction as of January 1, 2021, so that boundary is now a fixed, checkable fact rather than a moving one for any given city — but it still has to be checked per city, since it depends on that city's population and history.

None of that is optional detail for an AHFA application, because AHFA's own threshold list turns the zoning question into a document requirement, not a code lookup. The applicant must provide evidence that the property is properly zoned and consistent with the proposed use, in the form of a signed statement from the local jurisdiction — and AHFA will not treat the property as zoned if final approval is still contingent on further city meetings, approvals, or advertisement. For a site that sits outside any municipality — where no zoning applies at all — that threshold item becomes a different kind of evidence problem entirely, one the QAP text doesn't spell out in the same level of detail it gives the flood and environmental items.

The site file AHFA wants finished before you apply, not after

AHFA's threshold requirements are a completeness gate, not a scoring input: if any of them is missing, or materially fails to meet AHFA's defined standards, the application terminates during the completeness review, before the point-scoring evaluation ever runs. Three of those thresholds are the site file proper — the flood certification, the Environmental Site Assessment, and the market study — and all three have to be substantially finished at the time of application, not queued up for later in the process.

The Phase I ESA has to be prepared by an Environmental Professional to the current ASTM Phase I standard and satisfy the federal All Appropriate Inquiries rule at 40 C.F.R. Part 312. If certain time-sensitive AAI components run past 180 days before AHFA's review concludes, AHFA can require the report updated — at the applicant's expense. One genuinely useful piece of relief buried in AHFA's own environmental policy: for a vapor-intrusion condition, AHFA will accept a permanent passive vapor mitigation system installed under Alabama Administrative Code r. 335-15-4-.04 in lieu of ongoing compliance monitoring, rather than requiring active mitigation indefinitely.

Where Alabama compares favorably to California's own screening layer is the hazard data itself. The Alabama Department of Environmental Management runs a public e-Maps Portal and GIS Hub with a dedicated Brownfields inspector layer, downloadable in CSV, KML, GeoJSON, GeoTIFF, and PNG, with GeoServices, WMS, and WFS API access documented. That is a meaningfully more open posture than the 403s California's guide documents at DTSC's EnviroStor API and SWRCB's GeoTracker — though it says nothing, on its own, about whether every ADEM layer that matters to a given site (LUST, active landfills, voluntary cleanup sites) is complete or current, only that the access path itself is not walled off the way it is in some other states.

The sharpest trap in this whole phase isn't a data gap at all — it's a rule about the site itself. Any site change or alteration of any kind, or any change in the Ownership Entity, occurring after the application is submitted but before AHFA's approval, is a Negative Action that can terminate the application outright. Practically: whatever site goes into the application has to be the site, locked, through the entire review window — swapping to a marginally better parcel nearby after filing isn't a safe fallback, it's a way to lose the application entirely.

One project per county, a 70-point floor, and where the radius bites

AHFA's Point Scoring System ranks each application on Points Gained minus Points Lost. Points Gained has two top-level buckets — Project Characteristics and Applicant Characteristics — and Location is scored as a 10-point subcategory nested inside Project Characteristics, not a third parallel bucket.

Location is scored as a subcategory within Project Characteristics, worth exactly 10 of that bucket's 84 points, earned 2 points at a time for each of five listed neighborhood services — grocery store, pharmacy or drug store, convenience store, bank or credit union, hospital or doctor's office — found within 3 miles of the site by odometer distance, or within 5 miles if the site meets USDA Rural Development's 'Rural Area' definition for its Section 515/538 programs under 7 C.F.R. § 3560.11. On the other side of the ledger, negative-neighborhood-service deductions carry no maximum at all: 5 points off for a new project adjacent to any of thirteen listed incompatible uses (junk yards, processing plants, prisons, airports, and others), with 'adjacent' defined as nearby, not necessarily touching, and the list itself explicitly not exhaustive.

Above the individual score sits the allocation methodology that makes county geography, not just point totals, decisive. AHFA generally funds one project per county per cycle, after first funding the highest-scoring CHDO-owned project toward the mandatory 15% CHDO set-aside. Ties are broken by a nine-rung cascade: least aggregate participation by a single Responsible Owner, then HOME funding, then a census tract at or above 100% of county median family income, then the county with the fewest units in active or unplaced AHFA projects, then compliance history, then — notably — a project in a Qualified Census Tract supported by a Concerted Community Revitalization Plan approved by the local governing body within the prior 5 years (evidenced by a highlighted, 10-page-or-less excerpt from that plan), then intent for eventual tenant ownership, then least aggregate non-profit participation, and finally a public drawing held the next business day in AHFA's boardroom.

The 2-Mile Radius Requirement is a separate, earlier gate: AHFA will not consider a new-construction application, or a rehabilitation application for a project less than 50% occupied, if it sits within 2 miles — measured centroid-to-centroid by GIS — of any other AHFA-funded project from a prior cycle that hasn't yet reached 90% occupancy or been Placed-In-Service. Six exceptions apply, covering substantially occupied rehabs, Choice Neighborhoods/RAD/Capital Fund deals, National Register historic rehabs, PHA replacement housing, bond-financed pools of three or more existing properties, and paired bond/competitive applications on adjacent sites. The radius has to be affirmatively shown in the market study, not just checked and forgotten.

Two more site-adjacent items round out this phase. First, an appraised value that comes in below the sales-contract purchase price is itself an enumerated ground for AHFA to change or deny the credit allocation — Alabama's rough equivalent of California's tiebreaker discount for overpaying, except it surfaces after the application is already filed rather than as a modeled scoring penalty at LOI time. Second, don't conflate AHFA's own building-level Difficult Development Area designation — a discretionary tool under 26 U.S.C. § 42(d)(5)(B)(v), available only in narrow HOME-financing scenarios such as AHFA holding both first and second mortgages, or a prior AHFA HOME loan being repaid or extended — with the federal HUD-designated DDA map that drives the standard, self-executing 130% eligible-basis boost. They are two different mechanisms, triggered two different ways, and only one of them is a site-location question at all.