What is loss to lease in multifamily underwriting?

Loss to lease is the gap between the in-place rent on the rent roll and the rent the owner believes the same units could command today, shown here as a percentage of in-place rent (some shops divide by the market rent column instead, which gives a slightly smaller figure). It is an accounting measure of a gap, not a forecast that the gap gets collected. Whether any of it is collectible depends on what new leases and renewals are actually signing at in that submarket, which varies considerably by vintage and class.

How do you underwrite loss to lease when new-lease trade-outs are negative?

One approach separates the rent roll into the share of units likely to turn and the share likely to renew, then applies the trade-out spread observed in that market to each stream rather than moving every unit to the market rent column. Per RealPage's 2nd Quarter 2026 REIT Update, new-lease trade-outs at several large apartment owners ranged from roughly -5% to +1%, while renewal trade-outs reported in the owners' own 2Q 2026 releases ran roughly 3% to 5%, which puts most of the achievable capture on the renewal side. These are national operator averages and individual assets may differ materially.

Is the market rent column on a rent roll the same as achievable rent?

Usually not. The market rent column is typically an asking rent, and asking rent is gross of concessions, which averaged roughly 11% of asking rent among units receiving one in August 2026 per RealPage. A net effective comparison — asking rent less the amortized concession — tends to be the more defensible input, and concession practice varies widely by submarket.

Does a large loss to lease make a multifamily deal more attractive?

A large gap can indicate under-managed rents, or it can indicate a stale market rent column that no longer reflects what the submarket is signing. The distinguishing evidence is generally the last 90 days of signed leases at the property and at the comp set, not the gap itself. Deals priced on an uncollectible gap may be paying today for revenue that never arrives.

Multifamily Loss to Lease: Underwriting the Real Gap

Published October 22, 2026

A 7% loss to lease on a 200-unit rent roll looks like roughly $270,000 of annual revenue sitting in plain sight. Per RealPage's 2nd Quarter 2026 REIT Update, published August 2026, new-lease trade-outs over that quarter ran -0.7% at Equity Residential, -5.3% at Mid-America, and -3.3% at Camden Property Trust, against renewal trade-outs of roughly 3% to 5% at the owners that disclosed both, per those owners' own 2Q 2026 earnings releases. In a market where new leases sign below the expiring rent, the unit that turns does not move up to the market rent column — it may move down.

That is the mechanical problem with how loss to lease usually enters a model. The gap is real as arithmetic. Treating it as revenue that arrives on a turnover schedule is a separate assumption, and in 2026 it is the assumption doing most of the work.

Why Loss to Lease Is Not Free Upside

Loss to lease measures the distance between two columns on a rent roll: what residents pay now, and what the owner's software says the unit is worth. The second column is an asking rent. Asking rent is gross of concessions, and per RealPage's August 2026 concessions update, roughly 15% of stabilized units carried a concession averaging about 11% of asking rent — near the deepest reading in that series since it began in 2010. A unit advertised at $1,712 with six weeks free is signing closer to $1,515 on a net effective basis ($1,712 × 46/52 weeks).

The second problem is the schedule. Capture happens only when a lease is re-signed, so the pace is set by turnover and renewal pricing, not by the size of the gap. Those two streams have been moving in opposite directions: the resident who stays accepts roughly 3% to 5%, and the resident who leaves is replaced at a rent that may be flat to down. A model that applies the full gap to every expiring lease is implicitly assuming the new-lease stream behaves like the renewal stream.

What the 2026 Lease Data Shows

Sourced facts first, interpretation after.

Per RealPage's 2nd Quarter 2026 REIT Update (published August 4, 2026): new-lease trade-outs of -0.7% (Equity Residential), -5.3% (Mid-America), -3.3% (Camden), and +1.0% (Essex); blended trade-outs between -0.2% and +3.6%.
Per the owners' own 2Q 2026 earnings releases: renewal trade-outs of 5.2% at Equity Residential (July 22, 2026), 5.2% at Mid-America, and 2.8% at Camden (July 30, 2026).
Per the same RealPage update, Mid-America reported turnover of approximately 40% and Equity Residential a resident renewal rate near 60%.
Per RealPage's 3rd Quarter 2026 data update (published October 6, 2026): same-store effective asking rent growth of about 0.9% year over year, an average effective rent near $1,915, and occupancy averaging 95.4% in the quarter.
Per RealPage's August 2026 concessions update: concessions on approximately 15% of stabilized units, averaging about 11% of asking rent, up roughly 1.3 points year over year.
Per Yardi Matrix's September 2026 National Multifamily Report (press release dated October 7, 2026): a quarterly advertised-rent gain of 0.3%, year-to-date growth of 1.4%, and negative year-over-year growth in 14 of the 30 largest metros.

What this could mean, read as an interpretation rather than a forecast: the renewal stream is carrying nearly all of the rent growth in the data above, while the new-lease stream is roughly flat at best nationally and clearly negative in several large portfolios. If that pattern holds in a given submarket, the collectible portion of a loss-to-lease gap is a function of retention rather than turnover — which inverts the usual value-add logic, where turnover is the engine. These are national and operator-level averages, and individual submarkets, vintages, and classes vary substantially.

A Worked Example: A 7% Gap That Delivers Less Than Half

Illustrative figures throughout. A 200-unit Class B property in a secondary market. Average in-place rent of $1,600 per month, a market rent column at $1,712 — a 7% loss to lease. Annualized Gross Potential Rent (GPR) at in-place rents is approximately $3.84M. Assume 40% of units turn in year one and 60% renew, consistent with the turnover and retention figures above.

Year-1 revenue build	Method	Rental revenue
Offering memorandum pro forma	Full gap captured, plus 3% growth on top	~$4.23M
Gap captured at turnover	Turning units to the market rent column, renewals held flat	~$3.95M
Measured build	Renewals at +4.5%, new leases flat, concessions at the measured share	~$3.92M

The +4.5% renewal assumption sits toward the high end of the 2.8%–5.2% range the disclosing owners reported. At Camden's 2.8%, renewal revenue falls by about $39,000 (120 units × $1,600 × 12 × 1.7 points), and the measured build drops to roughly $3.88M.

The spread between the first and third rows is approximately $314,000 of year-one revenue. Carried down net of a 5% management fee, that is roughly $298,000 of Net Operating Income (NOI). Capitalized at 5.5%, it is about $5.4M of value — close to 12% of an illustrative $46M purchase price, or roughly $27,000 per unit of price that depends on a capture assumption rather than on operations.

The coverage effect is tighter than the value effect. On $29.9M of debt at a 65% Loan-to-Value (LTV) ratio, a 6.0% rate, and 30-year amortization, annual debt service is approximately $2.15M. A year-one NOI of $2.90M underwritten on the pro forma build is a Debt Service Coverage Ratio (DSCR) of roughly 1.35x; the same deal on the measured build is near 1.21x. Rate quotes and amortization terms vary by lender and vintage — verify current pricing directly with a lender.

The line item most often missed here is the concession tail. A concession granted on a 12-month lease signed in month two is still suppressing collected rent in month 13, so a year-one revenue build that nets concessions out of GPR and then layers a full-year rent increase on the renewal of that same lease has counted the recovery twice. Normalizing the rent roll against signed leases rather than against the market rent column is the same discipline that normalizing a messy T-12 demands on the expense side; for the revenue mechanics underneath it, see the Net Operating Income and T-12 statement references.

Where This Framing Breaks

The honest counter-case is that trade-out data from large public owners is a poor proxy for a specific Class B asset in a specific submarket. The portfolios cited above are concentrated in coastal and high-supply Sun Belt markets, and the negative new-lease prints are substantially a supply story — per RealPage, roughly 318,000 units were completed in the year ending 3Q 2026, which is heavy in absolute terms even at about 46% below the late-2024 peak. A property in a submarket with no competing lease-up may be signing new leases well above expiring rents while the national average is negative.

A genuinely under-managed rent roll is also a real phenomenon, not a modeling artifact. Where a prior owner held rents flat for several years out of inattention, the gap to achievable rent can be large and can be captured quickly, and refusing to underwrite any of it would mean systematically mispricing those deals. The test is evidentiary rather than theoretical: the last 90 days of signed leases at the property, and at three or four genuine comps, generally settle whether the market rent column is a target or a fiction.

There is a further limitation worth stating. The measured build above is more conservative than the pro forma, but it is not more certain — it substitutes one set of assumptions about retention and trade-out spreads for another, and retention itself responds to how aggressively renewals are priced. Pricing renewals at 5% to capture the gap tends to raise turnover, which pushes more units into the weaker new-lease stream. That feedback loop is difficult to model and is usually left out of both builds.

The Underwriting Takeaway

Loss to lease belongs in a model as two streams with separate spreads, not as one gap with a capture schedule. The renewal share carries a positive spread that recent data supports; the turnover share carries a spread that may be negative, and sizing it from the comp set's signed leases rather than from the rent roll's market rent column is the step that changes the answer.

A deal that only clears its coverage test on the full-gap build is a deal underwritten on national-average rent growth that the 2026 lease data does not currently show on the new-lease side. That position may still be the right one for a particular submarket. It is worth knowing it is the position being taken — the same two-speed caution that underwriting a bifurcated market and an occupancy-first revenue model apply to demand assumptions applies to the rent roll as well.

UWmatic is an AI-powered underwriting platform built for multifamily and mobile home park investors. Build the revenue line as separate renewal and new-lease streams and see what each trade-out spread does to NOI, coverage, and value. Try the free underwriting calculator →

This analysis reflects current market interpretations as of the publication date and may evolve as new data becomes available. Figures cited are drawn from public sources including Equity Residential's, Mid-America's, and Camden's 2nd Quarter 2026 earnings releases, RealPage Analytics' 2nd Quarter 2026 REIT Update, 3rd Quarter 2026 data update, and August 2026 concessions update, along with Yardi Matrix's September 2026 National Multifamily Report, and are subject to revision. Nothing in this post is investment advice; readers should conduct their own diligence and consult qualified professionals before making investment decisions.

Multifamily Loss to Lease: Underwriting the Real Gap