DSCR, LTV, and Debt Sizing for REPE Deals

Every REPE acquisition model eventually hits the same question: how much debt can this deal support? Lenders answer that with two constraints that show up in every IC memo, every broker's financing summary, and almost every superday technical: DSCR (debt service coverage ratio) and LTV (loan-to-value). Understanding what each one actually tests, and where candidates break them, is one of the highest-ROI topics in REPE interview prep.

DSCR: can the property's cash flow cover the debt?

DSCR = CFBD ÷ (Interest + Scheduled Principal)

CFBD (Cash Flow Before Debt) starts from NOI and subtracts capital items below NOI: tenant improvements, leasing commissions, capital reserves, and any other recurring or non-recurring capital outflows that reduce cash available to service debt. This is the critical distinction from NOI-based metrics: a deal can have strong NOI and still fail DSCR if TI/LC and reserves are heavy (common in value-add and lease-up assets).

The denominator is total debt service: cash interest plus any mandatory principal amortization. Some lenders test interest-only DSCR during an initial interest-only period; others require coverage on full P&I from day one.

A typical minimum DSCR for stabilized multifamily might be 1.20x–1.30x on a 10-year fixed-rate loan; value-add or transitional assets often require higher coverage (1.35x+) or accept a lower initial DSCR with a cash trap or lockbox until stabilization.

The annualization trap (again)

The single most common DSCR mistake in interview models: using annualized NOI when the property isn't actually stabilized. If you're underwriting a lease-up asset with 70% occupancy today but projecting 95% by month 18, the Year-1 CFBD is not (Stabilized NOI × 12). You need a month-by-month or at minimum a Year-1 average that reflects actual collections during lease-up. Overstating Year-1 CFBD makes DSCR look fine when a real lender would require reserves, higher equity, or a lower LTV.

LTV: how much of the purchase price is debt?

LTV = Loan Amount ÷ Value (or Total Cost Basis)

Value in the denominator is usually the lesser of appraised value and purchase price (the "lesser-of" test), though some lenders use total cost basis (purchase price + closing costs + initial capex). Always clarify which convention the lender uses; it matters when you're buying below appraised value or funding significant upfront capex.

Typical stabilized multifamily LTV ranges: 65%–75% for agency or life company debt; value-add deals often cap at 60%–65% LTV until stabilization, with a future supplemental loan or refi once the asset stabilizes.

LTV vs. LTC

LTC (loan-to-cost) uses total project cost (acquisition + capex + fees) as the denominator instead of value. Ground-up development and heavy value-add deals are sized on LTC, not LTV, because there's no stabilized value to capitalize until the business plan executes. A 70% LTC on a $100M total cost project means $70M of debt against $30M of equity — but if the stabilized value is only $110M, the exit LTV is 64%, not 70%.

How lenders actually size the loan

In practice, the loan amount is the minimum of what each constraint allows:

Max Loan (DSCR) = CFBD ÷ (Min DSCR × Debt Constant)
Max Loan (LTV)  = Max LTV × Value
Loan Amount     = MIN(Max Loan DSCR, Max Loan LTV, Lender Program Cap)

The debt constant converts a DSCR requirement into a maximum loan size: it's the ratio of annual debt service to loan amount at the assumed rate and amortization (e.g., a 5.5% rate, 30-year amort might produce a debt constant around 6.8%, meaning every $1M of loan requires ~$68K/year of debt service). If CFBD is $5M and the lender requires 1.25x DSCR with a 6.8% debt constant, max loan ≈ $5M ÷ (1.25 × 0.068) ≈ $59M.

Interviewers love asking you to walk through this math without a spreadsheet. Know the formula cold and be ready to explain which constraint binds (DSCR-limited vs. LTV-limited deals tell different stories about risk).

When DSCR and LTV conflict

A DSCR-limited deal means the property's cash flow, not its value, caps the leverage. Common on value-add assets where you're buying at a low cap rate (high price relative to in-place NOI) but the business plan drives NOI higher over time. The lender sizes on in-place or Year-1 CFBD, so you need more equity upfront even if the stabilized LTV would look fine.

An LTV-limited deal means the property's value, not its cash flow, caps leverage. Common on stabilized, high-occupancy assets with strong DSCR but where the lender's LTV cap (say 65%) is the binding constraint. These deals often support higher leverage at refi once you've executed the business plan and reappraised.

Knowing which constraint binds tells you where the real risk sits: cash flow execution (DSCR) vs. value/market risk (LTV).

Debt yield: the quick underwriting shortcut

Debt Yield = Year-1 NOI ÷ Loan Amount

Debt yield is a lender shorthand that ignores amortization and interest rate: it's the cap rate on the loan itself. A 9% debt yield on a $60M loan implies $5.4M of Year-1 NOI. Minimum debt yields vary by asset class and market (multifamily might be 7%–9%; office or retail higher). It's not a substitute for full DSCR analysis, but it's the first filter in many lender screens and shows up constantly in broker OM financing summaries.

Stress testing: what IC actually cares about

No real memo presents a single base-case DSCR. The standard stress cases:

• Rate stress: +100–200 bps on the refi or floating-rate assumption
• NOI stress: −5% to −10% on stabilized NOI (vacancy spike, expense overrun)
• Exit cap expansion: +25–50 bps on exit cap rate (reduces refi proceeds and exit equity)

A deal that passes 1.25x DSCR in the base case but breaks at 1.05x under a modest NOI haircut is a very different risk profile than one that still covers at 1.20x under stress. Interviewers want you to articulate this, not just recite the base-case number.

The interview checklist

When you finish any REPE model, verify:

1. CFBD (not NOI) is in the DSCR numerator
2. Year-1 CFBD reflects actual occupancy, not stabilized annualized income
3. LTV denominator matches lender convention (value vs. cost, lesser-of test)
4. You know which constraint binds and why
5. You've sensitized DSCR under at least one downside NOI case

Get these five right and you've covered the leverage section of almost every REPE technical. The rest is building the same bridge from GPR to NOI to value — just with debt layered on top.