Draft — not checked against a real transaction. The rent chain and the cash-flow structure are validated against figures the housing agencies publish; no closed deal has been reproduced in it. What has been validated, and how to check it against a deal you know

Why would a private company put money into affordable housing?

The Low-Income Housing Tax Credit (§42) is how most affordable housing in the US gets financed by private capital. A company invests equity in a project up front and in exchange receives federal tax credits over the next ten years — a dollar-for-dollar reduction of its federal income tax — plus the tax losses from depreciation. The return comes mostly from those tax benefits rather than rent: rents are capped, but demand at capped rents is deep, so occupancy is high and the income is steady. A firm that also builds has a second seat at the table — developer and contractor fees are earned on the construction itself, separate from the investment return.

The obligations are long. The credits pay out over ten years, the IRS compliance period runs fifteen, and the affordability restriction thirty: units must stay rent-capped and leased to income-qualified households the whole time. Fall out of compliance — or sell during the fifteen-year period — and the IRS takes back credits already claimed, with interest. So the money is illiquid, the cash yield is modest, and someone has to own compliance for a generation. This screener exists to test, before any of that is committed, whether a specific building can carry such a deal.

This is a screening estimate, not underwriting. It is a first-pass filter for whether a building is worth investigating further. It is not a pro forma, not a credit application, and not advice. Real feasibility depends on a physical survey, an environmental assessment, current allocation criteria, and a lender's own underwriting.

Start from a worked example

Each is a complete project you can load and then pull apart. The structures are typical and the arithmetic is exact; the buildings are invented. Change one thing at a time and watch what moves.

Inputs

Unit mix

How many homes, how big, and how deep the affordability goes. Every rent in the model is derived from this — the AMI column sets the ceiling, and the type column decides whether a home is capped at all.

Beds Units Avg SF Type AMI % Remove

Loading…

What has been validated, and what has not.

Checked against published agency figures. The rent chain is run against 15,540 figures published by the Illinois Housing Development Authority — 182 county schedules, every AMI band, every household size and bedroom count. IHDA publishes both the inputs (income limits) and the outputs (maximum rents), so the model is fed one and graded on the other by the agency that enforces the rules rather than by whoever wrote it. The cash flow's structure is checked against IHDA's PPA Workbook, the pro forma spreadsheet the agency makes applicants fill in — reserves inside total expenses, coverage measured after them, debt service converted monthly. The operating reserve basis is taken from IHDA's and Chicago's underwriting guides, which publish the same rule independently of each other, and the operating defaults trace to the City's published Underwriting Standards Guide. PMT, NPV and IRR are checked against Microsoft's published Excel worked examples, because no agency publishes an IRR to check against.

Not checked against a real transaction. No one has taken a deal that actually closed and reproduced it here from end to end. Everything above establishes that the model follows the published rules; none of it establishes that it matches what a real project did. The assumptions the rules do not fix — lease-up pace, cost escalation, exit pricing, how a particular lender sizes debt — are untested by any of it, and they are usually what decides whether a deal works.

So validate it for your own use before relying on it. Take a transaction whose outcome you know, ideally one you were part of. Enter comparable values — unit mix, AMI bands, acquisition and construction cost, rates and terms — and compare what comes out against what actually happened. Where the two diverge, and by how much, is the measure of whether this model fits your use case. That single comparison is worth more than every agency figure listed above.