Underwriting Section 8 Rent: The Vacancy and Collection-Loss Math

A wholesaler sends you a deal with "guaranteed government rent" in the subject line. The pro forma shows zero vacancy and zero collection loss because the Housing Assistance Payment arrives on the first of every month. The seller is technically describing something real, and the underwriting is still wrong.

The HAP portion of a Section 8 rent really is the most reliable income stream in residential real estate. But "reliable while the contract is active" and "no income risk" are different claims. Voucher rentals have their own loss lines, and they show up in different places than market-rate loss lines do. If you model a voucher unit with market-rate vacancy assumptions, you will be wrong in both directions at once: too pessimistic about occupied months, too optimistic about the gaps between tenancies.

This post walks through the three loss lines that belong in a Section 8 underwrite, then runs a full worked example on a Memphis single-family rental using FY2026 numbers.

What Is Actually Guaranteed, and What Is Not

Section 8 rent has two components, and they carry different risk.

The HAP payment is the portion the public housing authority pays you directly. While the HAP contract is in force and the unit passes inspection, this money shows up every month regardless of what is happening in the tenant's life. Job loss does not interrupt it. In a recession, the tenant's portion gets recalculated downward and the HAP portion grows to cover the difference. This is the real economic argument for the program, and it is legitimate.

The tenant portion is ordinary rent. The tenant pays roughly 30 percent of adjusted income, capped at 40 percent for new leases, and collecting it takes the same effort as collecting from any tenant. If the tenant portion is $290 a month, that $290 has normal collection risk.

And then there is the part the pro forma never shows: the income that does not exist yet. Between the day a voucher holder applies and the day the first HAP payment lands, there is a paperwork and inspection pipeline that market-rate rentals do not have. That pipeline is the largest single difference in the vacancy math.

Loss Line 1: The Lease-Up Gap

When you accept a voucher applicant, the sequence generally runs like this:

In an efficient PHA, that pipeline takes two to three weeks. In a backlogged one, it can take six to eight. Most PHAs will not pay for the days before the contract starts, so every week in the pipeline is a week of vacancy you carry. Some PHAs pay from the date the unit passes inspection, which helps, but you should confirm that with the specific agency rather than assume it.

For underwriting, I treat the initial voucher lease-up as one extra month of vacancy compared to a market tenant who signs and moves in the same week. On a turn between voucher tenants, the whole pipeline runs again.

The offsetting factor is tenure. Voucher tenants stay longer than market tenants on average, mostly because the voucher is tied to the unit until the tenant goes through a formal move process with the PHA. Where a C-class market rental might turn every 18 to 24 months, voucher tenancies commonly run several years. Fewer turns means fewer lease-up gaps, which is why the total economic vacancy on a voucher unit often ends up lower than a market unit even though each individual gap is longer.

The other thing that shortens the gap is marketing reach. The lease-up clock does not start until a qualified applicant finds you. Listing the unit where voucher holders are actually searching cuts the dead time in front of the RFTA. You can list a rental on VoucherMatch, which is built specifically for voucher household search, and shorten the front end of that pipeline.

Loss Line 2: Collection Loss on the Tenant Portion Only

Here is the mistake I see in both directions.

Optimists apply zero collection loss to the whole rent because "the government pays." Pessimists apply a fat 5 percent collection loss to the whole rent because they have heard voucher tenants are risky. Both are modeling the wrong base.

Collection risk lives almost entirely in the tenant portion. If the contract rent is $1,650 and the HAP is $1,360, then only $290 a month is exposed to nonpayment. Even a harsh assumption like 15 percent loss on the tenant portion works out to $43.50 a month, or about 2.6 percent of the total rent. The same 15 percent assumption applied to a fully tenant-paid market rent of $1,650 costs you $247.50 a month.

This asymmetry is the quiet advantage of the program and the reason the cash-flow comparison between Section 8 and market-rate tenants tilts the way it does in workforce neighborhoods. The income slice that can fail is small by construction.

Two cautions. First, the tenant portion is recalculated when household income changes, so the exposed slice moves around over the life of the tenancy. Second, if the tenant stops paying their portion, you still have to enforce the lease like any landlord, and the PHA will keep paying its share while you do. Model the loss, not a catastrophe.

Loss Line 3: Abatement Risk

The HAP payment has one kill switch: inspection compliance. If the unit fails its periodic inspection and you miss the repair deadline, the PHA abates the payment. Abatement means the HAP stops until the deficiency is fixed and re-verified, and in most programs those abated dollars are never paid back.

Since October 2025, HCV inspections run under NSPIRE standards, which put hard deadlines on health and safety items: 24 hours for life-threatening deficiencies, 30 or 60 days for the rest. I covered the inspection regime in detail in the Section 8 landlord math breakdown, so here I will just say what it means for the pro forma: a well-maintained unit with a responsive repair process should essentially never eat an abatement, and a deferred-maintenance unit can lose whole months of its largest income stream.

I budget abatement risk as a small allowance rather than pretending it is zero: about 1 percent of gross scheduled rent on a property I know is in good shape, more on anything I bought with inherited deferred maintenance.

The Worked Example: A Memphis 3-Bedroom

Memphis is a natural market for this math: deep voucher demand, single-family stock at workable prices, and a FY2026 3-bedroom Fair Market Rent of $1,683 for the Memphis, TN-MS-AR HUD Metro FMR Area. With the payment standard band at 90 to 110 percent of FMR, the local ceiling for a 3BR sits somewhere between $1,515 and $1,851 depending on where the PHA has set its standard. Memphis is also a Small Area FMR metro, so the operative number varies by ZIP; you can check the figures for a specific ZIP on VoucherMatch's Memphis Section 8 page.

Assume the deal looks like this:

The naive model takes $19,800 of gross rent, applies zero vacancy and zero collection loss, and produces this:

The realistic vacancy line assumes one tenant turn across a five-year hold with roughly ten weeks of total downtime: two weeks of make-ready, two to three weeks of marketing, and the RFTA-to-HAP pipeline on the back end. That is $4,125 of lost rent spread over five years, or $825 a year. The collection line is 10 percent loss on the $290 tenant portion. The abatement line is the 1 percent allowance.

Now the full picture:

With 25 percent down and a $104,250 loan at 7.0 percent over 30 years, annual debt service runs about $8,322. The naive model shows $3,194 a year of cash flow. The realistic model shows $1,633. Both are positive, but one of them is a number you can plan around and the other is a number a wholesaler made up.

Notice the compliance line item. Market-rate underwrites do not carry it. A voucher unit gets inspected on a schedule, and keeping a unit at inspection standard costs something even when nothing is broken: smoke and CO detector checks, handrail tightening, the window that will not latch. Three hundred dollars a year is my base assumption on a solid house.

The realistic model still clears 7 percent as a cap rate, which is the honest version of why this segment attracts cash-flow investors. The point is not that the voucher math fails. The point is that it survives honest assumptions, and you should use them.

How This Compares to a Market-Rate Underwrite

The same house rented at market to an unassisted tenant in the same neighborhood might command a similar rent with a different loss profile: faster lease-up (no RFTA pipeline), more frequent turns, and the entire rent exposed to collection risk. A typical C-class market underwrite in this rent band carries 7 to 9 percent combined vacancy and collection loss. The realistic voucher model above carries 7 percent, differently distributed.

That distribution matters more than the total. The voucher unit's losses are concentrated in predictable windows (turns) and controllable behaviors (maintenance), while the market unit's losses arrive whenever a tenant's finances break. If you have read the in-place Section 8 due diligence guide, you have seen the same principle from the acquisition side: verify the pieces individually instead of trusting the blended story.

Three Underwriting Mistakes to Avoid

1. Using the payment standard as the rent. The payment standard is a ceiling on subsidy, not your rent. Your rent has to clear rent reasonableness against comparable unassisted units. Underwrite the rent the comps support, and treat any gap up to the payment standard as headroom, not income.

2. Modeling zero vacancy. The HAP guarantee applies to occupied, contracted, passing months. It does not fill the gap between tenants, and that gap is structurally longer than a market turn. One extra month per turn is a fair base assumption.

3. Applying collection loss to the wrong base. Split the rent into HAP and tenant portions and apply loss only to the slice that carries it. This one change fixes most bad voucher pro formas, in both directions.

Running Your Own Numbers

The Section 8 rent analyzer pulls the FY2026 FMR and Small Area FMR for any ZIP code, along with the payment standard band and the local PHA, and the investment analyzer plugs those figures into a full deal model. For the general-purpose version of the math, the single-family calculator lets you set vacancy and loss assumptions line by line, which is exactly what this approach requires.

If you want to see what voucher demand looks like in a market before you commit to the model, the state and metro pages on VoucherMatch carry FMR, payment standard, and voucher program data for every county in the country. The Tennessee Section 8 page is where I would start for the Memphis example above.

Guaranteed rent is a real feature of this program. It is just not the same thing as guaranteed income, and the spread between those two ideas is exactly where the underwriting lives.