5 Common Mistakes in Rental Property Underwriting

These are not beginner errors. They show up in finished models, in offering memoranda, and in deals that closed. Each one systematically pushes the answer in the same direction — toward the deal looking better than it is — which is what makes them worth knowing individually.

1. Using the seller’s pro forma as the base case

The most common and the most expensive.

A pro forma is a marketing document. It is not fraudulent, usually, but every judgment call inside it was made by someone whose interest is a higher price. Those judgments compound.

What typically sits inside one:

• Vacancy at 0%, or at “actual” trailing occupancy from an unusually good year
• No property management fee, because the seller self-manages
• No capital reserve
• Property taxes at the seller’s assessed basis rather than yours
• Insurance at a premium bound years ago
• Market rents applied to units currently leased below market, with no cost or timeline for capturing the difference

None of those is a lie. Together they can move a stated cap rate by 150–200 basis points, which on a $600,000 building is well over $100,000 of implied value.

The fix is structural, not arithmetic. Do not open the seller’s numbers and adjust them — build the income statement independently from comps, quotes, and county records, and then compare. Anchoring is powerful enough that starting from their spreadsheet will pull your result toward theirs no matter how skeptical you intend to be.

Where the two versions diverge sharply, you have found your diligence questions. That is the actual value of the pro forma.

2. No capital reserve anywhere in the model

Roofs, HVAC systems, water heaters, sewer lines, parking lot surfaces, and windows all fail. None of them fail evenly across years, which is exactly why they get left out — there is no obvious annual line item for an event that happens once a decade.

So the model shows $4,000 of annual cash flow, and in year six a $22,000 capital event erases five years of it.

The convention that causes this is defensible on its own terms: capital expenditure sits below NOI, and market cap rates are quoted on NOI excluding reserves. Institutional owners exclude it from NOI and model it explicitly in the cash flow. Retail investors adopt the first half of that convention and skip the second.

The fix: carry a reserve line by name.

After the inspection, replace the rule of thumb with the actual schedule. If the roof has six years left and costs $18,000, that is $3,000 a year of known, dated capital — not a general reserve, a scheduled expense.

The test of whether you have done this properly: your model should show what happens to cash flow in the year the roof is replaced. If it does not, the reserve is decorative.

3. Ignoring tax reassessment on transfer

Jurisdiction-specific and frequently overlooked by out-of-state buyers, which is a large share of the people it affects.

In California, Proposition 13 caps annual assessment increases at 2% while an owner holds. A property held since 1998 may be assessed at a fraction of market value — and that assessment resets to your purchase price the moment you buy. The seller’s $4,200 tax bill becomes your $11,000 tax bill.

Florida’s Save Our Homes provisions and portability rules produce a similar discontinuity. Texas has its own mechanics and among the highest effective rates in the country. Many other states reassess on transfer to some degree.

An unmodeled $6,800 annual tax increase on a property producing $35,000 of NOI is a 19% reduction in NOI, and therefore roughly a 19% reduction in what the building is worth to you.

The fix: before making an offer, pull the parcel record from the county assessor, find the current assessed value and effective rate, and confirm the reassessment mechanic for that specific jurisdiction. Then apply the rate to your purchase price. This takes about ten minutes and is the highest-value ten minutes in the entire underwrite.

4. Growing rents without growing expenses

In multi-year models, this one hides well because each individual input looks reasonable.

Rent grows at 3% a year. Expenses grow at 2%, or at nothing at all, because the modeler set the rent growth assumption deliberately and let the expense rows inherit whatever the template had.

The result is silent margin expansion. Every year of the projection, NOI grows faster than revenue, and by year five the model shows an expense ratio the property has never achieved and no reason it should.

Real expense behavior over the past several years has run in the opposite direction. Insurance premiums in several states have grown faster than rents. Property taxes follow assessments, which follow prices. Labor-driven costs — management, maintenance, turnover — track wage growth.

The fix: grow expenses at least as fast as rents unless you can name the specific reason a line should lag. Then check the expense ratio in the final year of the projection against year one. If it has drifted down by more than a point or two, find out which row did it.

A related version: modeling rent growth at all in a market with flat or declining population. Rent growth is a market outcome, not a policy you set.

5. Confusing cash flow with return

The mistake that causes otherwise careful investors to buy the wrong deals.

Two properties:

Property A pays better today. Property B produces roughly twice the annualized return over a five-year hold, because rent growth compounds against fixed-rate debt service that does not, and because the exit value follows the NOI.

The reverse trap exists too, and it is worse. A deal with weak cash flow and a projected 18% IRR is entirely dependent on the exit assumption — and exit cap rate is the input that moves IRR more than any other. Property B is a good deal if the rent growth materializes. If it does not, the investor is holding a 1.6% yield and waiting.

The fix: never decide on one metric. Cash-on-cash tells you whether you survive year one; IRR tells you whether the hold was worth it; equity multiple tells you how much money actually came back. A deal that looks good on all three, under a stressed exit cap, is a deal. A deal that looks good on one is a bet on whichever assumption that metric happens to load on.

The pattern underneath all five

Every one of these errors moves the answer in the same direction.

Seller’s pro forma: optimistic. Missing reserves: optimistic. Unmodeled reassessment: optimistic. Expenses growing slower than rents: optimistic. Single-metric decisions: whichever metric flatters the deal.

None of them is random noise that averages out across a portfolio. They are biased in one direction, they stack, and an underwrite containing three of them can overstate value by twenty percent or more without a single line looking obviously wrong.

Which is why running the sequence the same way every time matters more than any individual judgment inside it. Consistency is what makes the comparison between two deals meaningful, and it is what keeps optimism from entering one property’s model and not another’s.

If you use software for this, the specific things to verify are the defaults: whether vacancy starts at zero, whether a management fee is included by default, whether there is a reserve line at all, whether expense growth is a separate input from rent growth, and whether the exit cap rate can be set independently of the entry cap. A tool that gets those defaults wrong will reproduce the seller’s pro forma faster than you could by hand. We checked how the main options handle these in best real estate deal analysis software.

FAQ

Which of these costs investors the most money? The seller’s pro forma, because it contains several of the others inside it. Tax reassessment is the largest single-line error in the states where it applies.

How much should I budget for capital expenditure? $200–500 per unit per year depending on age and condition, replaced by a specific schedule once you have an inspection report.

Is a 0% vacancy assumption ever appropriate? No. Even a fully leased building turns units, and turnover has both vacancy days and make-ready cost attached.

How do I know if my expense assumptions are realistic? Check the expense ratio against comparable properties. Small residential typically runs 35–50% of effective gross income. Materially below 30% means something is missing.