"Is this parcel worth an LOI?"
What "is this worth an LOI" actually requires
Colorado's allocating agency is the Colorado Housing and Finance Authority (CHFA), and its Qualified Allocation Plan (QAP) — currently the 2025-2026 QAP, Second Amendment, adopted by the CHFA Board on September 26, 2025 and approved by the Governor on November 13, 2025 — governs both the federal Low Income Housing Tax Credit and Colorado's own state Affordable Housing Tax Credit, created by HB 14-1017 and codified at Colorado Revised Statutes, Title 39, Article 22 (the QAP's own defined term, "the Colorado Act"). Before a Preliminary Application can even be filed, an Applicant must have submitted a Letter of Intent (LOI) — a formal, dated gate that doesn't exist in every state's process, and that falls roughly two months ahead of the Application deadline: for Round One 2026 (federal 9 percent credits), the LOI was due December 1, 2025 against a February 2, 2026 Application deadline; for Round Two 2026 (state credit plus noncompetitive federal 4 percent), the LOI was due June 1, 2026 against an August 3, 2026 Application deadline.
| Step | What it involves |
|---|---|
| Site control | Recorded deed/title commitment, or a fully executed agreement with evidence of the other party's ownership. A seller's Letter of Intent is explicitly not valid site control (Threshold #2). |
| Zoning | Pull zoning status directly from the local planning department, including parking requirements. For federal 9 percent or competitive state-credit deals, appropriate zoning must already be in place — not merely in process. |
| Site suitability | Proximity to employment, child care, schools, shopping, transit, medical services, parks; density fit; neighborhood character; slope, noise, hazards, floodplain, wetlands, utilities, power lines. |
| Census tract status | Qualified Census Tract (QCT), Difficult Development Area (DDA), and Small Area DDA lookups — federal HUD designations, same process as any state. |
| Income limits and rents | Pull the current-year CHFA rent-and-income-limits memo and check whether the county carries HERA Special limits or a hold-harmless carryover. |
| Environmental | Order a Phase I Environmental Report — required on every application, not just flagged sites, no older than 12 months. |
| Cost basis limit | Pull this year's RS Means-derived cost-basis-limit figures from inside the electronic Application itself; CHFA does not publish them as a standalone table. |
| Comparable awards | Read CHFA's most recent Award Report PDF — address, sponsor, unit count and credit amount only. |
| Prop 123 status | Check whether the site's jurisdiction has filed a Local Government Affordable Housing Commitment with DOLA — gates an entire layer of state financing. |
| Data type | How often it's actually refreshed |
|---|---|
| Site-screening data | Queried in seconds where a county's GIS holds up, phoned in where it doesn't |
| Underwriting data (income limits, rents) | Refreshed once a year by memo, on a fixed implementation date, with HERA and hold-harmless exceptions to track per county |
| Zoning data | Refreshed never, until a project gets bounced at Threshold #7 |
Two decisions come out of this phase, and the second one is sharper-edged than in most states. The first is whether to pursue the site at all. The second is whether zoning can plausibly be made "appropriate" and in place by the Application deadline — because unlike a documentation gap, zoning readiness is not on CHFA's one-day cure list. That list covers Threshold #1 (Minimum Score), #2 (Site Control), #12 (Energy Efficiency), #13 (Narrative) and #15 (Public Hearing, state credit only); Threshold #7 (Zoning and Entitlements) is not among them. A site that looks perfect on every other axis but can't clear zoning by the deadline doesn't get a second chance inside that round.
Parcel and zoning data: 64 counties, and the state says to call them
Colorado's Division of Property Taxation — the state agency inside the Department of Local Affairs (DOLA) that oversees how every county assessor values property — says plainly on its own site that it is not the source: "The Division of Property Taxation does not maintain individual property records and does not accept property tax payments... To look up your property assessment and other property-specific information, please contact your local county assessor." There is no statewide parcel dataset standing in for that, and no CHFA-published equivalent of it either. Colorado's 64 counties each run their own assessor and GIS stack, and a screening tool has to treat that as 64 separate integration problems, not one.
The Denver metro area has a regional planning body — the Denver Regional Council of Governments (DRCOG) — the way Southern California has SCAG, and DRCOG's Regional Data Catalog (data.drcog.org) is described as an open repository covering transportation, demographics, housing and land use for the region. What was not found in this pass is a DRCOG dataset with the shape of SCAG's zoning layer: a parcel-keyed join field, a normalized zone-code vocabulary, a vintage field per jurisdiction. That absence should be stated plainly rather than assumed away — a Colorado screening product should not budget for a SCAG-equivalent showing up ready to ingest, and should verify DRCOG's actual holdings directly before building against them.
Zoning itself carries the same structural problem it does everywhere: a zone code is a pointer into a municipal or county code, not a value, and Colorado has no statewide normalization layer to resolve it. CHFA's own Threshold #7 requires zoning status documentation straight from the local planning department — not from any third-party GIS layer — which is itself a signal about how far an ingested zoning dataset can be trusted at application time rather than merely at screening time.
Site suitability is a judgment call — until zoning becomes a hard gate
This is the largest structural difference from California's screen, and it should be visible in the product rather than papered over. Where CTCAC scores site amenities on a fixed table with a negative-points regime for a wrong certified distance, CHFA's site-suitability factors sit inside "Criteria for Approval" — the same discretionary bucket that covers market conditions, project readiness and financial feasibility — not inside the numeric Scoring Criteria that produces the 130/115/95-point threshold. The QAP's own language: "Sites will be evaluated based on suitability and overall marketability, including, but not limited to, proximity to employment, child care, schools, shopping, public transportation, medical services, and parks/playgrounds, applicable amenities and access to services for the target population; maximizing the allowable housing density; conformance with neighborhood character and current land use patterns supporting residential development; and slope, noise (e.g., railroad tracks, freeways), environmental hazards, flood plain, wetland issues, utilities, and power lines." No per-factor point value is attached, and the surrounding section reserves CHFA the right to reject an application "in CHFA's sole opinion" on market and readiness grounds — the same discretionary standard that governs site suitability.
That means a Colorado screening tool can surface every one of those factors as evidence — walk-adjacent transit, hazard-layer hits, density headroom, neighborhood context — but none of it computes to a number the way a CTCAC amenity table does. Only a narrow set of location-related items actually carry a fixed, addable point value in the 2025-2026 QAP.
| Item | Points | QAP section |
|---|---|---|
| Existing/planned TOD site — half-mile walk to a transit corridor with easy job-center access; density should be maximized | 3 | Section 5.B.2.b |
| Community affordable housing priority location — must show the project fits a documented community need | 3 | Section 5.B.2.a |
| QCT project contributing to a Community/Concerted Revitalization Plan | 1 | Section 5.A.3 |
| Non-metro county, population 180,000 or fewer | 5 | Sections 2.A and 5.B.3.b |
Those four items add up to at most 12 points toward a 130-point federal 9 percent threshold. Everything else about site quality rides on the discretionary Criteria for Approval review, not the scored table.
The harder gate sits upstream of scoring entirely: appropriate zoning must already be in place for any federal 9 percent or competitive state-credit application. That is a materially different risk than California's operative site-control threshold, where zoning and entitlement risk is folded into a scored "project readiness" narrative rather than enforced as a hard pass/fail line before the round even runs.
The Phase I is universal — and the hazard layers behind it are only partly open
Unlike agencies that require a Phase I Environmental Report only once a red flag turns up, CHFA's Threshold #8 makes it mandatory on every application, covering every parcel in the proposed site, no older than 12 months from the Application date (extendable to 24 months only case-by-case, and only if the Phase I reported no Recognizable Environmental Conditions). If RECs turn up, a Phase II has to be submitted with the same application — there is no "flag it and resolve it after award" path.
The federal baseline underneath that report is the same as any state: FEMA's National Flood Hazard Layer is public and unauthenticated. Wildfire — the hazard that actually threatens Front Range and mountain-corridor sites the way flood threatens sites elsewhere — sits behind Colorado's own portal, run by the Colorado State Forest Service; the current front door is the Colorado Forest Atlas, an Esri ArcGIS-based viewer at coloradoforestatlas.org (the older "coloradowildfirerisk.com"/COWRAP branding now redirects there). Whether the underlying wildfire-risk data is available as a queryable REST service the way CAL FIRE's Fire Hazard Severity Zones are on data.ca.gov, or only as a rendered map layer inside the viewer, was not confirmed in this pass — that distinction changes what a screening product can actually build, and should be verified against the live service rather than assumed.
Two Colorado-specific interactions are worth flagging directly. First, CHFA's own discretionary basis boost — up to 30 percent, awarded when CHFA determines a project needs it to be economically feasible — is explicitly unavailable to a project that already qualifies for the federal 30 percent boost through a QCT, DDA or Small Area DDA, and it never applies to acquisition basis or to federal 4 percent projects. A screen that flags a QCT hit and separately flags "eligible for CHFA basis boost" without checking that exclusion is double-counting a benefit that doesn't stack. Second, every Housing Tax Credit and CHFA-loan project in Colorado is "held harmless" against a rent-limit decrease, and 37 of Colorado's 64 counties carried HERA Special (pre-2008 placed-in-service) limits in 2026 alongside the standard table — a screen that reads a single current-year number per county understates achievable rent on a meaningful share of Colorado deals.
Where the screen stops: cost, comparable awards, and the Prop 123 layer
Construction cost has the same hole it does everywhere, with a Colorado-specific wrinkle. CHFA's Method Three cost-basis-limit calculation is built on RS Means-derived standards, revised annually — but unlike CTCAC's published PDF of dollar thresholds by county and bedroom count, the current year's Colorado limits live inside the electronic Application spreadsheet itself, not as a standalone public table. There is nothing to screen against before pulling this year's Application file directly from CHFA's site.
Comparable awards are thinner than California's. CHFA's Award Reports are single-page PDFs listing project name, address, city, unit count, sponsor and credit amounts — no APN, no census tract, no unit-by-AMI-band breakdown. The 2026 Round One report, for example, lists 14 projects totaling 634 units, $23,003,741 in federal 9 percent credit, $7,000,000 in standard state credit and $494,000 in Transit-Oriented Community (TOC) credit as of May 19, 2026 — real, usable numbers, but with nothing to join against a parcel table the way CTCAC's 61-column List_of_Projects.xlsx allows in California.
The genuinely Colorado-only layer — and the one most likely to silently kill an otherwise-strong site — is Proposition 123. Colorado voters passed it in November 2022, dedicating 0.1 percent of state income tax revenue to a State Affordable Housing Fund, now codified at Colorado Revised Statutes, Title 29, Article 32. Sixty percent flows to an Affordable Housing Financing Fund managed by the Office of Economic Development and International Trade (OEDIT) and administered by CHFA (Land Banking, Equity and Concessionary Debt programs); 40 percent flows to DOLA's Division of Housing as an Affordable Housing Support Fund. None of the OEDIT/CHFA-administered funding is available for a project in a jurisdiction that has not filed a Proposition 123 Local Government Affordable Housing Commitment with DOLA. Those filings run on an annual November 1 cycle with escalating unit-growth targets — 9 percent over three years, 6 percent over two, or 3 percent over one, for the first cycle — and the next milestone lands squarely inside the life of the current QAP: second-cycle commitment filings were due November 1, 2026, with local-government compliance on unit count and "Fast Track" permitting-policy implementation due December 31, 2026. A site in a jurisdiction that never filed, or that has fallen out of compliance, is invisible to that entire financing layer regardless of how clean the parcel and zoning data look.
Land Banking itself is specific enough to underwrite against directly: grants to local and tribal governments, and forgivable loans to qualifying nonprofits, capped at the lesser of $5 million or CHFA's own concluded appraised value; a 2 percent deferred interest rate; a requirement to have proper zoning and a development plan within 5 years and to be permitted and funded within 10; and a mandatory environmental report on any transaction over $750,000.
One more layer is worth tracking rather than screening on yet. The Transit-Oriented Communities Act (HB24-1313, codified at C.R.S. § 29-37-201 et seq.) sets a housing-opportunity-goal formula — 40 dwelling units per acre multiplied by a jurisdiction's transit-area acreage (§ 29-37-204) — but it only binds a local government that qualifies as a "transit-oriented community": wholly or partly inside a metropolitan planning organization, population of 4,000 or more, and at least 75 acres of transit area (§ 29-37-202(11)). Qualifying jurisdictions had to submit a housing-opportunity-goal report by December 31, 2026, with full certified compliance due December 31, 2027. Until a jurisdiction actually certifies, treating proximity to a bus route as a live density entitlement is the same overreach the California guide flags for HQTA-derived AB 2011 signals — a proximity fact, not yet a zoning fact.
The final honest limit is the same shape it is in every state: whether the county planner actually fast-tracks the project, whether the jurisdiction's Prop 123 filing survives its own compliance check, whether a stronger 9 percent deal files in the same statewide round — none of that is a dataset, and none of it should be synthesized. What a Colorado screen can and should settle before the LOI: HERA- and hold-harmless-adjusted rents for the county, whether zoning is genuinely in place rather than merely likely, current Phase I status, and whether the jurisdiction carries a live Proposition 123 commitment.
Where this goes wrong
- Treating a seller's Letter of Intent as site control. Threshold #2 requires a recorded deed/title commitment or a fully executed agreement with evidence of the other party's ownership; the QAP states explicitly that Letters of Intent are not valid forms of site control.
- Screening a site as viable for a federal 9 percent or competitive state-credit round without confirming zoning is actually in place. The QAP requires appropriate zoning in place for these applications, and zoning readiness is not on the one-day cure list that covers Thresholds #1, #2, #12, #13 and #15 — miss it and there's no fixing it before the deadline. (Threshold #7's formal QAP name is "Readiness-to-proceed"; "Zoning and Entitlements" is the specific documentation requirement inside it.)
- Missing the Letter of Intent deadline and assuming there's still time to apply. LOIs fall roughly two months ahead of the Application deadline (Round One 2026: LOI December 1, 2025; Application February 2, 2026) — no LOI, no application in that round.
- Skipping the Phase I Environmental Report because a site 'looks clean.' Threshold #8 requires a Phase I on every application regardless of red flags, no older than 12 months (24 months only if that Phase I found no Recognizable Environmental Conditions).
- Modeling Colorado like a regional-apportionment state. CHFA states it will not consider requests for set-asides beyond the Code-required 10 percent nonprofit set-aside — there is no California-style per-region credit pool to check availability against; the only geographic lever is a scoring priority for non-metro counties of 180,000 population or fewer.
- Double-counting the CHFA discretionary basis boost with a QCT/DDA/SADDA hit. The CHFA basis boost is explicitly unavailable to a project that already qualifies for the federal 30 percent boost through a Qualified Census Tract, Difficult Development Area or Small Area DDA, and it never applies to acquisition basis.
- Reading only the current-year AMI column for a county. Thirty-seven Colorado counties carried HERA Special (pre-2008) limits in 2026 alongside standard limits, and every Housing Tax Credit and CHFA-loan project is separately held harmless against any decrease — a screen showing one number per AMI tier per county is showing the wrong one for a meaningful share of Colorado deals.
- Assuming Colorado has a statewide parcel or zoning layer analogous to California's SCAG regional dataset. Colorado's Division of Property Taxation states plainly that it does not maintain individual property records and directs users to the county assessor; no equivalent of SCAG's parcel-keyed, normalized zoning schema was confirmed for the Denver region or the state as a whole.
- Treating CHFA's cost-basis-limit figures as a public, standalone table the way CTCAC's threshold basis limits are. CHFA's Method Three limits live inside the current year's electronic Application spreadsheet, not a separately published PDF — there is nothing to check before pulling this year's file.
- Using CHFA's Award Report as a parcel-joinable comparable-awards dataset. It is a single-page PDF with address, sponsor and credit amount — no APN, no census tract, no AMI-mix columns — unlike CTCAC's structured List_of_Projects.xlsx.
- Assuming a jurisdiction is eligible for Proposition 123 Land Banking, Equity or Concessionary Debt funding without checking its Local Government Affordable Housing Commitment status. Eligibility is gated on the jurisdiction having filed with DOLA's Division of Housing, filings run on an annual November 1 cycle, and the jurisdiction has to stay in compliance — an otherwise-excellent site in a jurisdiction that hasn't filed, or has fallen out of compliance, is invisible to that entire funding layer.
- Treating proximity to a bus route as a live density entitlement under the Transit-Oriented Communities Act before the jurisdiction has actually certified. HB24-1313 only binds jurisdictions that qualify (inside an MPO, population 4,000+, 75+ acres of transit area) and only once they clear housing-opportunity-goal reporting (due December 31, 2026) and full certification (due December 31, 2027).
- Collapsing Colorado's discretionary Site Suitability review into a scorable checklist. The QAP's site factors — proximity to services, density fit, hazards, floodplain, noise — sit inside Criteria for Approval, evaluated at CHFA's sole discretion, not inside the fixed-point Scoring Criteria; a tool that assigns point values to them is inventing a scoring system CHFA doesn't publish.
- HUD
- LIHTC
- State QAPs
- IRS § 42
- Housing Finance Agencies