LTV vs LTC: When Each Constraint Governs Debt Sizing: Stabilized, Construction, and Value-Add

Key Takeaways

Institutional CRE LTV/LTC vs Other Meanings

Before going further: this article is about loan to value (LTV) and loan to cost (LTC) as the collateral-side pair of institutional sizing constraints in commercial real estate: CMBS conduit, life insurance company, agency Fannie/Freddie multifamily, commercial bank, bridge, and HUD 221(d)(4) construction.

A few disambiguations worth getting out of the way: "LTV" in consumer-mortgage glossary content refers to a residential single-family loan over the home's appraised value: the arithmetic is the same, the underwriting box (FICO, DTI, PMI triggers above 80% LTV) is not. "LTC" outside CRE is polluted by Long-Term Care insurance. Hard-money fix-and-flip LTV/LTC content uses similar vocabulary but at retail-scale advance rates against single-asset SFR, not the institutional register.

Why LTV and LTC Are Different Tests

The cleanest way to hold the distinction: LTV asks what the asset is worth; LTC asks what was spent. An LTV constraint is a value test: the lender protects against the asset's appraised value falling below the loan balance. An LTC constraint is a cost test: the lender protects against having lent more than the verifiable dollars that went into the project.

The reason both tests exist is that during construction and value-add transition, value and cost diverge. A half-built apartment building has no operating value: no rent roll, no NOI, no stabilized appraisal. The only verifiable number is what the sponsor has spent. LTC is the lender's discipline against that reality: lend a fraction of the documented cost basis because there is nothing else to lend against. Once the project stabilizes, cost becomes irrelevant; the asset has a defensible appraised value derived from comparable sales and income capitalization. LTV is the discipline against that reality.

A mental shorthand: at the same advance rate, whichever base is smaller sets the loan. If stabilized value is below total project cost, the LTV test produces the smaller loan and governs; if value is above cost, the LTC test does. LTC is the operative test during construction and value-add transition for a different reason: there is no stabilized value to test yet, so cost is the only verifiable base. At stabilization, the property has an appraised value and LTV becomes the operative test. The institutional lifecycle moves through both regimes; the worked examples below walk a value-add deal and a ground-up development through both binding constraints.

LTV: The Stabilized Constraint

The loan to value ratio is the lender's collateral-coverage discipline on an income-producing asset. The formula is the cleanest in institutional CRE:

LTV = Loan Balance ÷ Appraised Value

The lender publishes a maximum LTV (65%, 70%, 75%) and the borrower's loan proceeds are capped at that fraction of appraised value. At 70% LTV against a $50M appraised stabilized value, the maximum loan is $35M. Push appraised value down by 20% (an illustrative cap-rate-expansion scenario on the same property) and the same 70% advance rate sizes to $28M, a $7M proceeds compression on the same deal, before any change to the headline LTV maximum.

The denominator, appraised value, is where institutional practice gets subtle. Three distinct valuation regimes apply, producing three different numbers for the same property:

• As-is value. Current condition, occupancy, and operating state. Used by permanent lenders on stabilized acquisitions and refinancings: the appraiser derives value from in-place rent roll, comparable sales, and income capitalization. The most conservative regime; gives no credit for prospective lease-up, capital improvements, or projected NOI growth.

• As-stabilized value. The projected value at stabilized occupancy and stabilized NOI under the sponsor's business plan. Used by bridge and value-add lenders to size the takeout test. Larger than as-is on a value-add deal; a lender may haircut the sponsor's projection to build cushion (the examples below assume an illustrative 10% haircut).

• As-completed value. The projected value at construction completion (improvements built per spec, but before lease-up to stabilized occupancy). Used by construction lenders as a pre-stabilization test. Sits between as-is (zero or land-only on ground-up) and as-stabilized (assumes occupancy is achieved).

A lender quote of "70% LTV" without specifying which V is half-information. 70% LTV against as-is for a stabilized acquisition is one number; 70% LTV against as-stabilized for a value-add takeout is a different (higher) number; 70% LTV against as-completed for a construction loan is a third. Every term sheet must specify the valuation regime: sponsors who don't ask consistently misread the proceeds picture.

Illustrative LTV planning ranges by lender type as of Q1 2026 (most sit about 5 points below the 2021 levels shown; agency is unchanged):

Illustrative LTV planning ranges by lender type as of Q1 2026, with 2021 levels for comparison; not quoted lender terms. The valuation regime (as-is, as-stabilized, as-completed) determines what "V" is in the formula. Source for the agency row: Freddie Mac Optigo Fixed-Rate Loans term sheet.

LTC: The Construction / Value-Add Constraint

The loan to cost ratio is the lender's collateral-coverage discipline on a project that has no stabilized operating value to lend against. The formula:

LTC = Loan Balance ÷ Total Project Cost

Total project cost (TDC) is the verifiable dollar basis of the project: land acquisition, hard cost (the GMP contract with the GC), soft cost (architectural, engineering, legal, permits, lender fees, title), and reserves (construction interest reserve, operating deficit reserve at completion, contingency). Critically, TDCexcludes equity-side closing costs that don't flow into the project: sponsor acquisition fees paid back to the GP, equity placement fees, and organizational expenses. Get the denominator right and the LTC calculation falls out cleanly; get it wrong and the sponsor consistently overshoots proceeds.

Illustrative LTC planning ranges by lender type as of Q1 2026:

Illustrative LTC planning ranges by lender type as of Q1 2026; not quoted lender terms. HUD 221(d)(4) is a government-insured program whose advance rate depends on program tier; confirm the current limit in the HUD MAP Guide before modeling. The agency forward row is an LTV and DCR takeout test, not a loan-to-cost advance. Source for the forward row: Freddie Mac Conventional Forwards term sheet.

The lender-type LTC range is wider than the LTV range because the underwriting box is more sponsor- and market-specific. A regional bank lending to a long-relationship sponsor on multifamily construction in a primary market underwrites to the high end of its range (75%). The same bank lending to a first-time sponsor on speculative office construction in a tertiary market underwrites to the low end (65% or lower, with completion guaranty and additional reserves). HUD 221(d)(4) is the outlier: the program insures multifamily construction loans at a program-specific advance rate, but the closing process is long and the mortgage insurance premium and prevailing wage requirements add cost. Those costs raise the equity dollars and carry the sponsor must plan for; they do not reduce the proceeds the advance rate supports.

The Constraint Hierarchy

LTV and LTC are the collateral-side pair of a four-constraint sizing framework. The full stack is:

• LTV (value test). Loan ÷ appraised value. Governs stabilized acquisitions and refinancings.

• LTC (cost test). Loan ÷ total project cost. Governs construction and value-add transition.

• DSCR (cash-flow test). NOI ÷ annual debt service. Governs when income is the binding variable: rate-stressed refinancings, thin-coverage assets. Covered inDSCR: Sizing Constraints, Amortizing vs IO.

• Debt yield (yield test). NOI ÷ loan balance. Often the lead test on CMBS conduit sizing; binds on premium-priced assets where cap rate is below the lender's debt yield threshold. Covered inDebt Yield: Why Institutional Lenders Prefer It.

On any given deal, the tests a lender applies run in parallel and the lender takes the minimum. Different lenders weight the constraints differently because of how their balance sheets and bond-rating frameworks are constructed:

Three patterns are visible in the matrix. First, CMBS leads with debt yield because the rating-agency framework keys off NOI ÷ loan balance, the cash-yield discipline that proved more durable through the GFC than DSCR (which is gameable through coupon and amortization elections). Second, life co and bank permanent debt lead with LTV and DSCR jointly: the value test and the cash-flow test running together, with the lower binding. Third, construction debt (bank construction, HUD 221(d)(4)) leads with LTC because the asset has no operating value to lend against. Bridge debt is the unique case that runs both LTC and LTV: LTC at peak draw during transition, LTV against the as-stabilized appraisal as the takeout test.

When LTC Governs: A Worked Value-Add Example

Work the math on a value-add multifamily deal in a primary market. The sponsor is acquiring a 1985-vintage 240-unit garden-style property at $30M, with a planned $5M renovation program (in-unit upgrades, common-area amenities, exterior facade refresh). Total project cost: $30M acquisition + $5M renovation = $35M TDC. Current as-is appraised value: $30M (the acquisition price; the lender's appraiser confirms). Sponsor's projected as-stabilized value at the end of the 18-month renovation: $40M (the value-creation thesis). The exit lender is assumed to haircut that projection by 10%, to $36M (illustrative).

The sponsor is choosing between two financing structures: a bridge loan at 75% LTC sized against TDC, or a permanent loan at 70% LTV sized against as-is value with the renovation financed out-of-pocket.

Worked example: value-add multifamily, $30M acquisition + $5M reno = $35M TDC, $30M as-is value, $40M sponsor-projected as-stabilized value ($36M after an illustrative 10% lender haircut). The 80% as-is cap on the bridge's initial advance is an illustrative assumption. The takeout sized on the haircut value ($25.2M) falls $1.05M short of the $26.25M bridge commitment.

The bridge commitment at 75% LTC ($26.25M) is $5.25M more than the permanent at 70% LTV against as-is ($21M). The LTC test counts the $5M renovation spend as part of the cost basis, which the permanent lender sizing only on as-is does not, because the planned improvements don't yet exist. The bridge also carries an as-is LTV test (80% in this illustration), so the first advance is capped at $24.0M ($30M × 80%) and the remaining $2.25M funds as renovation holdback as the work is completed. LTC governs the total commitment during the value-add transition. At stabilization (month 18+), the takeout is sized on LTV and the bridge converts out.

This is the lifecycle in microcosm: LTC sizes the bridge at $26.25M, and LTV sizes the exit at $25.2M after a 10% haircut ($40M × 90% × 70%). The takeout falls $1.05M short of the bridge, so the sponsor funds the difference with equity. If the lender haircuts the projection by 25% instead, the takeout is $21.0M ($40M × 75% × 70%) and the shortfall is $5.25M. A takeout modeled on the sponsor's unhaircut $40M would show $28.0M and a $1.75M lift, which is the number a sponsor sees when it skips the haircut. The lender's discipline shifted from a cost-protection regime (during the work) to a value-protection regime (after the work). The sponsor that models only LTV would have under-financed the renovation phase; the sponsor that models only LTC would miss the LTV-driven shortfall at the takeout.

When LTV Governs: A Worked Stabilized Refinance

Move to the opposite end of the lifecycle. A stabilized core-plus office property in a primary market was acquired in 2016 at $50M with $35M of 10-year fixed-rate CMBS debt (70% LTV at origination). The loan matures in 2026, so the sponsor must refinance the $35M balance into a new permanent loan. Current as-is appraised value: $40M (cap-rate expansion since 2022 has left value 20% below the $50M purchase price). NOI is roughly flat at $2.4M.

Sized against the four constraints at current market terms (life co quote, 65% LTV, 1.30x DSCR, 9% debt yield, 6.00% fixed coupon, 30-year amortization):

• LTV constraint: $40M as-is value × 65% = $26.0M max loan
• LTC constraint: Not applicable (stabilized refinance, no project cost; lender does not apply LTC on permanent stabilized debt)
• DSCR constraint: $2.4M NOI ÷ 1.30x = $1.85M max debt service. At a 6.00% coupon and 30-year amortization the loan constant is about 7.19%: $1.85M ÷ 7.19% = $25.7M max loan
• Debt yield constraint: $2.4M NOI ÷ 9.0% = $26.7M max loan

The lower-of binding constraint among the three applicable tests is DSCR at $25.7M, with LTV ($26.0M) a close second. The same loan balance ($35M) that sized at 70% LTV against $50M of value at origination now exceeds all three applicable constraints simultaneously. The refinance gap, the gap between the current balance and the maximum refinanced proceeds, is $9.3M.

The mechanical driver is the LTV step-down. At origination, the loan sat at the LTV limit (70% × $50M = $35M). In 2026, the same 70% advance rate against $40M would size to $28M; the tightened 65% advance rate sizes to $26M. The loan has not changed; the appraised value has dropped 20%. The LTV constraint became binding mechanically, before any consideration of DSCR or debt yield. This is the central pattern in a refinance wave driven by value compression (see Refinancing Risk: Maturity Wave and Default Rate Sensitivityfor the full cluster treatment).

The Construction-to-Perm Transition

The institutional discipline that ties LTV and LTC together is the construction-to-perm transition: the moment when a project's collateral underwriting shifts from cost-protection (LTC) to value-protection (LTV). It's a literal closing event: one loan paying off another, two lenders applying two tests against two denominators on the same property. A canonical ground-up multifamily timeline:

• Month 0 (construction closing). Bank construction lender commits $56M at 70% LTC against projected $80M TDC. Initial draw funds land and early hard cost.

• Months 1–24 (draw period). Loan funds incrementally against documented hard-cost spend. Balance rises to peak draw of $56M (70% LTC). As-completed value tracks to roughly $85M per the construction appraiser, but the asset is not yet stabilized.

• Months 24–36 (lease-up). Construction complete; project enters lease-up. The construction loan converts to mini-perm or interest-only conversion, with the conversion test requiring a target leased percentage and a target DSCR.

• Month 36 (stabilized takeout). Permanent lender sizes to 65–70% LTV against the in-place appraised value of the now-stabilized property, say $95M. The 70% LTV permanent loan at $66.5M takes out the $56M construction balance, returning $10.5M of equity: the "lift" from successful construction. Here value ($95M) is above cost ($80M), so the LTC test ($56M) is the lower of the two during the draw period.

The gap risk is cost overrun or value shortfall. A cost overrun on a fixed loan commitment raises the sponsor's equity requirement; LTC itself rises only if the lender funds the overrun with debt (a 20% overrun funded entirely with debt takes 70% LTC to 75%). A value shortfall (the as-stabilized appraisal comes in below projection) shrinks the LTV-based takeout. Forward-commitment programs such asFreddie Mac's Conventional Forwardsprovide takeout certainty to developers and construction lenders during construction, with a forward commitment term of up to 48 months; bank construction lenders may also require a takeout commitment letter from a permanent lender before they close. LTV and LTC are the two poles between which the lifecycle moves.

2026 Cycle: How Both Tests Squeeze Proceeds

Two forces now press on both collateral tests at once: lower appraised values tighten LTV, and higher construction costs widen the equity requirement under LTC. Three drivers:

Cap-rate expansion lowers appraised values. On flat NOI, value moves inversely with the cap rate. For illustration, a multifamily cap rate moving from 4.0% to 5.5% implies value compression of about 27%, and a move from 4.0% to 6.0% implies about 33%. NOI growth offsets part of that compression, so realized value declines on properties with growing NOI are smaller than the cap-rate arithmetic alone. The mechanical consequence: an LTV constraint that was non-binding at origination can bind at refinance on the same loan balance.

Construction costs have risen. Turner's Building Cost Index annual average rose from 1,199 in 2021 to 1,426 in 2024, about 19%, and reached 1,485 in 2025, about 24% above 2021 (see the Turner Building Cost Index releases in Sources). When cost grows faster than the value a project creates, the two tests pull apart: the same advance rate on a larger cost base sizes a larger LTC loan, while the as-stabilized LTV loan stays tied to value, so the LTV test tightens and the sponsor's equity requirement widens. A cost overrun on a fixed commitment raises equity as well; LTC itself rises only if the lender funds the overrun with debt (a 20% overrun funded entirely with debt takes 70% LTC to 75%).

The 2026 maturity wave is mechanically an LTV-step-down event. A loan balance stays flat while the value in the denominator moves with cap rates, so loans originated against higher appraised values face a higher implied LTV at maturity. The refinance gap is, at its core, an LTV-step-down problem driven by value compression rather than underwriting error at origination.

How to Model It

A useful LTV/LTC sizing model needs to do five things: compute the four-constraint cross-check on every term sheet, explicitly model the three valuation regimes, walk the construction-to-perm lifecycle with LTC at peak draw and LTV at stabilized takeout, cross-reference against the lender's maximum advance rate by lender type, and flag the binding constraint explicitly so the IC memo doesn't reverse-engineer it.

• 1. Build the TDC schedule. Land + hard cost + soft cost + reserves + financing cost. Exclude equity-side closing costs. Reconcile to the construction lender's approved budget line by line; where the sponsor's TDC differs from the lender's, size to the lender's number.

• 2. Build the appraisal regime by phase. As-is at acquisition. As-completed at construction completion. As-stabilized at the sponsor's stabilization date (with a lender haircut at the exit test), then the in-place value for the permanent takeout. Keep each number on a single tab.

• 3. Run the four-constraint cross-check. Compute loan balance at each constraint the lender applies. The minimum is the binding loan. Output should identify which binds (illustrative: "LTV at $37.7M binds, tighter than DSCR at $42.0M and debt yield at $44.0M").

• 4. Run the lifecycle walk. For development and value-add, model peak construction draw against TDC (LTC binding) and stabilized takeout against as-stabilized value (LTV binding). The gap or lift between the two is what the construction lender needs to see refinance them out.

• 5. Compare term sheets on the same basis. A CMBS quote at 70% LTV may size lower than a life co quote at 65% LTV if debt yield binds tighter on CMBS. A HUD-insured construction quote with a higher advance rate also carries mortgage insurance premium and prevailing wage cost, which raise project cost and the equity dollars required. Normalize all quotes, then compare on proceeds, blended cost of capital, prepayment flexibility, and recourse.

Common Mistakes

Seven errors we see repeatedly on LTV/LTC sizing: each one shows up in IC memos that misread the binding constraint:

• Treating LTV and LTC as substitutes for each other. They are not interchangeable. LTV protects against value collapse on a stabilized asset; LTC protects against cost-basis overhang on a non-stabilized project. They apply to different stages of the lifecycle, and the institutional discipline is to run both in parallel during transitional phases: the lower binds.

• Using stabilized value when the lender uses as-is. The most common LTV misread. The sponsor quotes the year-5 stabilized value to the permanent lender; the lender sizes against today's as-is appraisal, which is materially lower. The proceeds delta is the entire value-creation thesis, not proceeds the permanent lender will fund at closing. Always specify which V the lender is using.

• Including equity-side closing costs in the LTC denominator. Sponsor acquisition fees paid back to the GP, equity placement fees, organizational expenses: these do not flow into protective collateral for the lender and are excluded from TDC for LTC purposes. Including them inflates the LTC denominator and overstates the loan the LTC constraint supports.

• Forgetting the lower-of rule. The single biggest sizing error. Sponsors model the LTV constraint, see a $35M proceeds figure, and quote that to the IC. The DSCR constraint may be tighter at $28M; the debt yield constraint may be tighter at $30M. The binding loan is the minimum across the tests the lender applies. Modeling any single constraint in isolation systematically overstates proceeds.

• Assuming the headline LTV maximum applies to every property type. For illustration, a 70% LTV ceiling at a CMBS conduit might be the headline for stabilized multifamily and industrial, while office might be sized at 60–65%. Tertiary-market assets can get further haircuts. The headline number is a starting point; the binding constraint by property and market is often lower.

• Quoting LTV against the broker's appraisal estimate rather than the lender's. The broker's BPO and the lender's appraisal can differ materially. The lender's appraisal is the one that sizes the loan. Always re-test LTV against an institutional appraisal basis; a 10% value difference on a $40M property is $4.0M of value, or $2.8M of proceeds at 70% LTV, that a sponsor who skips this step will find at closing.

• Missing the LTV step-down at refinance. A loan that originated at 65% LTV against its origination-date value does not refinance at 65% LTV today: the V has moved against the loan. Sponsors who model the refinance at the original LTV without re-running the appraisal at current cap rates systematically miss the refinance gap. A refinance gap in a falling-value market is mechanically built out of this error.

Related Articles

LTV and LTC are the collateral-side pair of the four-constraint institutional debt-analysis cluster. Sibling deep-dives:

• DSCR: Sizing Constraints, Amortizing vs IO: the cash-flow leg of the four-constraint stack. The IO vs amortizing proceeds lift that pairs with LTC sizing on bridge debt.
• Debt Yield: Why Institutional Lenders Prefer It: the yield leg of the four-constraint stack. The CMBS pivot and why debt yield often binds first on conduit deals where LTV optically clears.
• Interest Rate Risk: Caps, Swaps, Hedging: the cap economics that protect against the rate-stress that pushes DSCR through the LTV constraint on floating-rate paper.
• Prepayment Analysis: Defeasance and Yield Maintenance: the prepayment cost of refinancing into a lower-LTV environment when the rate has moved against you.
• Refinancing Risk and the Maturity Wave: the 2026 context where the LTV step-down at refinance can be the binding constraint.

FAQ

Sources

Reference materials behind this article:

• Freddie Mac. Optigo Fixed-Rate Loans term sheet (maximum LTV and minimum DCR by term and payment type).
• Freddie Mac. Optigo Conventional Forwards term sheet (forward commitments for new construction; 80% maximum LTV, 1.25x minimum DCR, forward commitment term up to 48 months).
• Turner Construction Company. Turner Building Cost Index, 2024 Fourth Quarter and 2026 First Quarter (annual average index values for 2021 through 2025).