Underwriting red flags in an offering memorandum

Almost nothing in a good offering memorandum is false. That is what makes it effective. The rents are real rents somebody somewhere is paying. The expenses are real expenses somebody actually incurred. The cap rate is real arithmetic on real inputs. Every individual statement survives scrutiny. It is the selection that does the work, and selection is invisible unless you know what was left out.

So this is not a piece about catching liars. It is a piece about the twelve places where the omission usually lives, what each one is worth in dollars, and which document settles it.

The dollar figures below run against the 24-unit deal in the companion piece, how to underwrite a multifamily deal: $3,000,000 asking price, $346,068 of effective gross income, $178,165 of real NOI against the broker's stated $248,923. Six of the twelve flags below account for that entire $70,758 gap, which is $1,132,128 of value at a 6.25% cap. On a $3,000,000 deal.

1. Pro forma rents wearing in-place clothes

The tell. A rent roll exhibit with a column headed "Market" or "Achievable," and a revenue summary that uses that column without a loss-to-lease line anywhere beneath it. Sometimes the column is labeled honestly and the summary quietly uses it anyway.

Why it works. Loss to lease is the least intuitive line in real estate finance, so its absence does not look like an absence. It looks like a simpler statement.

How to catch it. Multiply the rent roll by twelve. Then find the rental revenue line in the T-12. If the rent roll annualizes to $358,800 and the revenue summary starts at $372,960, you have found $14,160 that nobody is paying. Do this before you read another page. It takes ninety seconds and it reframes the whole package.

What it is worth here. $14,160 of phantom revenue, about $227,000 of value at a 6.25% cap, which is 7.6% of the asking price.

A related move is subtler and worth watching for: the rent roll is real and current, but three units were re-leased last month at a new, higher rent after a renovation, and those three become "the new market rent" applied to all twenty-four. Three data points from the best units in the building are not a market. Ask for the last twelve signed leases, all of them, in order.

2. The T-12 that is secretly a T-3

The tell. A revenue line labeled "T-3 annualized" sitting in the same column as expenses labeled "T-12." Or a footnote reading "trailing three months annualized to reflect current operations."

Why it works. It is presented as a courtesy, a more current picture. And sometimes it genuinely is: a property that finished lease-up in March should not be valued on January's occupancy.

The problem is the mismatch. Revenue gets the flattering recent window; expenses get the full year, or worse, the full year from before the insurance renewal. Nothing is fabricated. The two halves of the income statement are simply measured over different periods, and the difference lands entirely in NOI.

How to catch it. Demand the T-12 month by month, all twelve columns, never a single annual total. Then do three things:

• Annualize each of the last three months individually. If December is 15% above the T-12 and January and February are not, December is not a trend, it is a month with a large one-time reimbursement in it.
• Look at the expense rows across the twelve columns. Snow removal, turnover, and landscaping are seasonal. A T-3 built on October through December in a northern market carries almost no snow cost and almost no summer landscaping.
• Check whether repairs and maintenance is suspiciously smooth. Real repairs and maintenance is lumpy. Perfectly even monthly numbers usually mean somebody accrued an estimate rather than recording actual invoices.

What it is worth. Varies, and that is the point. A T-3 that overstates NOI by 12% on this deal is $21,380 of NOI and roughly $342,000 of value. A mismatch this cheap to create should never survive contact with a buyer who asks for twelve columns.

The mirror-image version is the T-1: one month annualized, always the best month, usually labeled "current run rate." Treat "run rate" as a phrase that means "we picked the month."

3. Property taxes frozen at the seller's basis

This is the single most expensive item on the list, and it is not even hidden. It is just left alone.

The tell. A property tax line that is obviously low relative to the asking price. On this deal, $31,500 of taxes against a $3,000,000 price is 1.05% of value. The effective rate in the jurisdiction is 1.60%.

Why it works. The number is entirely accurate. It is what the seller paid last year. Nobody claimed it is what you will pay.

How to catch it. Pull the assessor's record yourself, which is public and takes about four minutes. Find the current assessed value, the effective millage, and, most importantly, the reassessment rule on transfer. Then run: your tax = purchase price x effective rate = $3,000,000 x 1.60% = $48,000

The rules vary enormously and the variation is real money. Some states reassess to sale price on transfer. Some cap annual increases but reset on sale. A few reassess on a multi-year cycle regardless of transfer, and some offer abatements that survive or die with the sale. In appeal-friendly jurisdictions a buyer can often win back part of the increase, but underwrite the full reassessment and treat any appeal as upside.

What it is worth here. $16,500 of NOI, $264,000 of value. That is 8.8% of the asking price, sitting in one line, in plain sight.

4. The expense ratio with no management, no payroll, and no reserves

The tell. An expense ratio in the low forties, or a per-unit operating expense number under $5,000 on an eighties-vintage property. Both should stop you cold.

Why it works. Because the seller genuinely does not pay these. He self-manages, he does the maintenance himself on Saturdays, and he funds capital out of pocket when something breaks rather than reserving for it. The statement is a true record of his costs. It is a poor forecast of yours.

How to catch it. Three lines, always, no exceptions.

Management. Third-party management on a small deal runs 4% to 8% of EGI, often with a per-unit monthly floor that governs on smaller properties. A package showing 2% or 3% is showing you an institutional rate on a 24-unit that no institutional manager would take. Charge the fee even if you plan to self-manage, because your buyer at exit will.

Payroll. If there is a maintenance function, there is a payroll cost. If the seller does it himself, price the replacement: a part-time burdened tech is roughly $18,000 on this deal. If you plan to contract it out instead, payroll goes to zero and repairs and maintenance rises to absorb it. What is not available is neither.

Replacement reserves. Almost no broker statement carries them, and a defensible argument exists that they belong below NOI for valuation, because market cap rates are quoted on NOI before reserves. Fine. But then they have to appear in the cash flow, and they have to appear in the lender's sizing, because agency and most bank lenders size debt on net cash flow, which is NOI less an underwritten reserve. A package that omits reserves from the valuation and the cash flow and the return table has removed them from the universe.

What it is worth here. Payroll at $18,000 and management at 3% instead of 5% total $24,014 of NOI, or roughly $384,000 of value. Reserves of $7,200 a year are not a valuation item but they are $36,000 of real cash across a five-year hold, and they are the difference between a DSCR of 1.32x and one the lender will not fund.

5. Occupancy quoted without saying which occupancy

The tell. A single occupancy percentage on the front page, large, with no qualification. "96% occupied."

Why it works. Everyone knows what occupancy means, which is exactly the problem, because there are two of them.

Physical occupancy is units with a body in them divided by total units. Economic occupancy is rent actually collected divided by rent theoretically collectible. The gap between them is where concessions, delinquency, employee units, and down units live.

A property can be 96% physically occupied and 84% economically occupied, and the second number is the one that pays your mortgage.

How to catch it. Compute it yourself from the two documents you already have: economic occupancy = T-12 collected rental revenue / (rent roll GPR x 12)

Then check three specific things on the rent roll:

• The balances-owed column. If four tenants owe more than one month, you have a collections problem that the occupancy number is hiding. Ask for a delinquency aging.
• Concessions. Look for units leased at a rent matching market with a note about free weeks. A twelve-month lease with six weeks free is an 11.5% rent reduction that never appears as vacancy.
• Non-revenue units. The model unit, the manager's unit, the down unit awaiting a flood remediation. These are often shown as occupied and produce nothing.

What it is worth here. Underwriting 5.0% vacancy with no concession and no credit loss line, against a real 6.0%, 1.0% and 1.5%, is $13,054 of NOI and about $209,000 of value.

6. Other income that has not happened yet

The tell. A line called "Ancillary income opportunity," or an other-income total that exceeds the T-12's by a suspiciously round amount.

Why it works. These are small numbers individually, they are plausible, and they are tedious to check. $28,000 versus $18,970 does not feel like an argument worth having.

At a 6.25% cap, that $3,030 difference on this deal is $48,480. It is an argument worth having.

How to catch it. Tie every other-income line to a T-12 row. Then interrogate the three that are most often projected rather than collected:

• RUBS. Ratio utility billing is the most common upside claim in multifamily, and it is genuinely available in many markets. It is also restricted or prohibited in some jurisdictions, requires lease language you may not have until renewal, and takes a full lease cycle to roll in. If it is not being billed today, it is a business plan. Put it in an upside scenario, not in year one.
• Pet rent and fees. Verify against the rent roll's pet column. "We could charge pet rent" is not income.
• Parking. If parking is currently free and included in the lease, converting it is a renewal-cycle project with real resident attrition attached.

Anything you cannot tie to a T-12 row or a signed contract goes in the upside case with a label on it.

7. "Recently renovated"

The tell. Photographs of a leasing office and one unit interior, the phrase "over $400,000 invested in capital improvements since 2019," and no schedule of what that money bought.

Why it works. It is a true statement about money spent. It says nothing about what was deferred. A seller preparing for sale spends on what photographs well: paint, landscaping, signage, a clubhouse, new flooring in the turned units. What does not photograph is the roof, the parking lot, the sewer lateral, the electrical panels, and the sixty-year-old cast iron stack.

How to catch it. Ask for the capital expenditure detail by year and by category, and read it for what is missing rather than what is there. Then get a property condition assessment and give it to an engineer, not to yourself.

Five items produce most of the surprises: roofs, paving, plumbing risers and laterals, electrical service and panels, and windows. Get a remaining-useful-life estimate on each.

Then put the number in your basis, not in a footnote:

Forty-four basis points of yield, gone, before you have collected a dollar. The offering memorandum will never show you the second number. It is the only one that describes your investment.

8. The "market" cap rate

The tell. "Priced at a 6.5% cap, below the submarket average of 6.0%."

Why it works. A cap rate is a ratio, and a ratio has two inputs, and the sentence tells you about neither. Three separate things can be doing the work.

The numerator. If the quoted cap rate is computed on pro forma NOI rather than in-place NOI, it is not comparable to anything. This is the most common version by a wide margin, and it is usually disclosed in a footnote in six-point type.

The denominator. Cap rate on price ignores closing costs and immediate capital, which is flag 7 above.

The comparison set. "Submarket average" is a phrase that requires a source. Ask which sales, on what dates, at what vintage, with what unit mix and what occupancy at closing. A 2019-vintage lease-up trading at a 5.5% cap tells you nothing about a 1984 walk-up with deferred plumbing, and it is a perfectly honest data point to include in an average.

How to catch it. Recompute it yourself on your own NOI and your own basis, then ignore the quoted number entirely. And build your comp set from sales you can verify, with the four attributes that actually drive comparability: vintage, unit mix, submarket, and occupancy at close. Price per unit is the laziest comp metric in existence, because it treats a 650-square-foot one-bedroom and an 1,150-square-foot three-bedroom as the same thing. Use price per square foot alongside it, and rent per square foot alongside that.

9. Year 1 in the offering memorandum is your Year 3

The tell. A five-year projection whose Year 1 already includes stabilized occupancy, full renovation premiums on every unit, and the RUBS program fully rolled in.

Why it works. Time is invisible in a table. A column labeled "Year 1" carries no information about how long the work took, and the reader supplies an optimistic default.

How to catch it. Build the ramp yourself, and be honest about the physics:

• Renovations require vacancy. You cannot renovate an occupied unit, so the pace is capped by natural turnover unless you are paying people to leave. On a 24-unit with 50% annual turnover, an all-units interior program is a two-year project at best.
• Renovated units carry downtime. Two to four weeks per unit of lost rent, which a Year 1 column almost never carries.
• RUBS and fee increases roll in at renewal, so a full lease cycle passes before the revenue is fully in place.
• The premium is achieved on a lease, which is signed by a person, in a market, at a moment. Three signed leases at the renovated rent in that submarket in the last six months beat any rent survey ever produced.

Then compare their Year 1 to your Year 3. If they match, the projection was not a projection. It was a description of a finished state with an early date on it.

10. Insurance quoted off an expiring policy

The tell. An insurance line that has not moved in three years on a statement where everything else has.

Why it works. It is a bookkeeping fact. The seller's premium is priced off the seller's loss history, the seller's deductible, and a policy bound in a different market.

How to catch it. Get a real quote. Not an estimate, not a per-unit rule of thumb, an actual quote from a broker who has seen the loss runs and knows the deductible structure you intend to carry. Ask for the loss runs in diligence, because a property with two water claims in three years prices differently from one with none.

This line has moved more than any other operating expense in recent years in wind, hail, and wildfire exposed states, and in some markets it has moved enough on its own to change what a property is worth. Inheriting the seller's number is the fastest way to be wrong by six figures of value. A $6,000 miss on a 24-unit is $96,000.

11. The lease expiration cliff

The tell. Nothing. This one is never called out, because it is not an income statement item. It lives in two columns of the rent roll that most people skim past.

How to catch it. Sort the rent roll by lease expiration date and count by month. Three patterns matter.

The cliff. If nine of twenty-four leases expire in the same sixty-day window, you have concentrated renewal risk into one season, and if that season is your market's worst leasing quarter, you will fill those units with concessions. Staggering expirations is a year of asset management work.

The month-to-month tail. Month-to-month tenants are portrayed as flexibility and are usually the opposite: they are long-tenured residents at rents well below market who have declined to sign at a new rate, and they will leave rather than renew at market. A rent roll with 30% month-to-month is a rent roll with a turnover event queued up.

The lease-up cohort. If a third of the leases were signed in the last ninety days at rents that look strong, find out what concessions were given. Newly signed leases at market rent with six weeks free are a rent roll that will reprice downward at renewal.

12. The return table with debt that does not exist

The tell. A returns page showing a 17% IRR, with financing assumptions in a footnote: "Assumes 75% LTC, 5.75% fixed, five years interest-only, 1.20x DSCR."

Why it works. Nobody checks the terms against what a lender will actually quote this week, and the package was probably written six weeks ago.

How to catch it. Call a lender, or three, and price it for real. Then check four things in the footnote:

• The rate. Compare to today's quotes for this asset type, size, and market.
• The proceeds. Run all three tests: LTV, DSCR, and debt yield. The package will show one, and it will be the one that produces the largest loan. In a higher-rate market coverage usually binds well before value does, and the equity check is bigger than the LTV headline suggests.
• Interest-only. Full-term interest-only flatters cash-on-cash and the IRR, does nothing for the asset, and leaves the entire principal balance outstanding at exit. It is a financing choice presented as performance.
• Floating rate. If the assumption is floating, the rate cap has a cost, the cap has a term shorter than the loan, and replacing it is a real cash outlay that the return table almost never carries.

And check whether the exit cap rate is below the going-in. Selling at a tighter cap than you bought is not an assumption, it is a wish, and on most deals it is quietly responsible for a third of the projected return.

Two more that are not numbers

"Below replacement cost." Nearly always true and nearly always irrelevant. Existing buildings trade below replacement cost for long stretches, and a building's value is the present value of its cash flows, not what it would cost to rebuild it. Replacement cost is a ceiling on new supply, which matters for your rent growth assumption in a specific and limited way. It is not a valuation argument.

"Call for offers, Thursday." Deadline pressure is a legitimate process tool and it is also the oldest way in the world to shorten diligence. If the timeline does not allow you to pull the assessor's record, get an insurance quote, and read twelve columns of a T-12, the timeline is telling you what the seller expects a buyer to skip. Bid on your own timeline and say why. Sellers take offers from buyers who close.

The two-hour first pass

Before you build anything, run this sequence. It kills most deals and it costs you an afternoon rather than a week.

1. Annualize the rent roll. Compare to the T-12 rental revenue line. Explain the gap.
2. Compute economic occupancy from the T-12 and the rent roll. Compare to the headline.
3. Pull the assessor's record. Reset the tax line to your purchase price.
4. Add management, payroll, and reserves if they are missing.
5. Recompute NOI. Recompute the cap rate on your NOI and on your basis, not on the price.
6. Sort the rent roll by expiration. Look for the cliff and the month-to-month tail.
7. Size the debt three ways. Take the lowest.
8. Now decide whether the deal deserves a full model.

Six of those eight steps are arithmetic on documents you already have. The reason they get skipped is not difficulty. It is that step 1 alone means keying twenty-four rows off a PDF, and step 2 means keying twelve months of a scanned operating statement, and by the time you are done you have spent the afternoon transcribing instead of thinking.

That is the part worth handing off. Altyst reads the rent roll and the operating statement together and ties them against each other, so a rent roll that does not support the T-12's revenue line comes back flagged, with the document each figure came from, instead of flowing quietly into your NOI. Steps 1 through 7 are done before you have finished reading the cover page.

What this is really about

None of these twelve items make a deal bad. Every one of them makes a deal differently priced, and pricing is the entire job.

The best acquisitions people are not the ones with the sharpest instincts. They are the ones who ask for the tax bill, the loss runs, the twelve columns, and the last twelve signed leases, every single time, on every single deal, including the ones they like. The discipline is boring, which is exactly why it is an edge.

And when you do walk, walk with a number. "We are at $2,600,000, and here is the line-by-line bridge from your $248,923 to our $178,165." Brokers do not respect low offers. They respect low offers with a rent roll analysis attached, and they remember who sent one when the deal comes back six weeks later at a different price.

Underwrite any property in seconds. Then go argue about the assumptions, which is the part that was always the actual work.

Illustrative figures throughout, tied to the worked example in how to underwrite a multifamily deal. Tax rules, insurance markets, and utility billing regulations vary by jurisdiction and should be confirmed locally. Educational content, not investment, tax, or legal advice.