The Underwriting Question Every Lender Must Answer
A middle-market sponsor delivers a Confidential Information Memorandum for a $40 million first-lien term loan on a manufacturing business. The CIM reports $10.2 million in Adjusted EBITDA: a 34% lift above the $7.6 million reported on the audited income statement. The proposed leverage multiple is 3.9x. On paper, it clears the threshold. But does it clear underwriting?
This is the central challenge of EBITDA normalization: every dollar of an add-back directly inflates the denominator in the Debt/EBITDA ratio, compresses the apparent leverage, and shifts credit risk onto the lender. A rejected $1 million add-back at a 4x multiple does not simply reduce EBITDA by $1 million; it effectively reduces the supportable loan quantum by $4 million. The consequences are asymmetric and the stakes are permanent once the credit agreement is signed.
This matters more than ever in the current environment. The gap between reported and adjusted EBITDA is now a front-line underwriting signal. As credit markets move through a more disciplined cycle, diligence teams are no longer accepting seller-provided normalization schedules at face value. Lenders are independently rebuilding the earnings bridge, and the scrutiny is justified by the data.
The Analytical Framework: Building the Lender's Own EBITDA Bridge
Institutional credit underwriting treats the seller's Adjusted EBITDA as a starting point, not a conclusion. The lender's job is to reconstruct a Lender-Underwritten EBITDA, a figure that reflects only sustainable, recurring, operating cash flows that will service debt across the full loan tenor.
The framework operates in four stages:
- Anchor to Audited or Tax-Return Reported EBITDA. Begin with the trailing twelve months (LTM) of EBITDA derived from audited financial statements or, for smaller borrowers, tax returns. Run-rate figures based on partial-year extrapolation must be scrutinized with particular care.
- Walk the Add-Back Schedule Line by Line. Demand documentary support for every adjustment. Categorize each add-back as (a) clearly acceptable, (b) conditionally acceptable with haircut, or (c) rejected.
- Apply the Lender's Own Haircuts. Even accepted add-backs may require probability-weighting or phase-in schedules, particularly for synergies and cost savings not yet in the run-rate.
- Rebuild Key Credit Metrics from Lender EBITDA. Recalculate Debt/EBITDA, Interest Coverage Ratio (ICR), and Fixed Charge Coverage Ratio (FCCR) on the adjusted base before sizing or approving the loan.
Working Example: A $10M EBITDA Manufacturing Company
| Line Item | Seller's Schedule ($M) | Lender's Assessment ($M) |
|---|---|---|
| Reported EBITDA (LTM, Audited) | $7.60 | $7.60 |
| + One-time legal settlement | $0.80 | $0.80 ✓ (documented) |
| + CEO compensation above market | $0.60 | $0.30 ✓ (haircut to market rate delta) |
| + Facility relocation costs | $0.45 | $0.45 ✓ (non-recurring, evidenced) |
| + Stock-based compensation (SBC) | $0.50 | $0.00 ✗ (rejected, real economic cost) |
| + Projected cost synergies (not realized) | $0.25 | $0.00 ✗ (forward-looking, not in run-rate) |
| Adjusted EBITDA (Lender-Underwritten) | $10.20M | $9.15M |
The $1.05 million delta is not cosmetic. At the proposed 3.9x leverage on seller EBITDA, the $40M loan maps to 4.37x on lender EBITDA: a materially different risk profile and one that may breach internal credit policy or require additional equity cushion.
The Analysis: Calculating Credit Metrics and Stress-Testing
With lender-underwritten EBITDA established at $9.15 million, the underwriting team rebuilds all three core credit metrics:
Core Credit Metric Calculations
- Net Leverage Ratio: ($40M debt − cash) / $9.15M EBITDA = 4.2x–4.4x (depending on liquidity). This compares to a typical direct lending covenant of ~4.0–5.0x for senior secured middle market transactions.
- Interest Coverage Ratio (ICR): $9.15M EBITDA / interest expense. At SOFR + 500bps (approximately 9.8% all-in as of mid-2026), annual interest on $40M = ~$3.92M. ICR = 2.34x: at the low end of the acceptable range for senior secured direct lending.
- Fixed Charge Coverage Ratio (FCCR): (EBITDA − CapEx − Taxes) / (Interest + Required Amortization). With $1.5M normalized CapEx and $0.75M in taxes, FCCR drops below 1.5x, a standard minimum covenant threshold, and warrants a closer look.
Stress Scenario
A conservative underwriter runs a 20% EBITDA stress scenario (representing a moderate recession or volume compression). At $7.32M stressed EBITDA:
- Net Leverage spikes to 5.3x–5.5x: outside typical covenant headroom.
- ICR compresses to 1.87x: approaching the 1.75x–2.0x floor commonly set in maintenance covenants.
- FCCR breaches below 1.0x, signaling inability to cover all fixed charges from operating cash flow.
This stress output informs both deal structure (does the lender need more equity, better covenants, or a smaller loan?) and covenant design (what triggers early warning and intervention rights?).
Industry Standards: What Institutional Lenders Are Seeing in 2025–2026
The current credit environment provides a clear empirical backdrop for why rigorous EBITDA underwriting is non-negotiable.
Market Data on Normalization Discipline
The market has structurally tightened its standards on add-backs. The shift in lender posture is measurable and documented across multiple data sources:
- According to practice data from 2025–2026, the typical reported-to-adjusted EBITDA gap runs 20–30% of reported earnings in lower middle market transactions, and wider gaps are a primary red flag signaling aggressive normalization that diligence teams will challenge.
- SBC add-backs have been effectively eliminated from sponsor underwriting models in 2026. Buyers and lenders alike now routinely exclude stock-based compensation from the earnings base, regardless of its treatment in the seller's QoE, as it represents a real economic cost to the business.
- Due diligence providers commonly disallow 10–30% of the EBITDA add-backs a seller proposes, with recurring expenses misclassified as one-time charges being the most frequently rejected category.
- Synergy add-backs in credit agreements are increasingly permissible only if realized within 18 months of closing and reviewed by an independent accounting firm, a structural discipline that was largely absent in the 2021–2022 vintage.
📊 Data Box: Private Credit Default & Recovery Benchmarks (2025–2026)
- Private credit default rate (2025): Moody's estimated the range at 1.6%–4.7%, depending on whether distressed exchanges are included. Fitch, applying a broader definition, reported the U.S. private credit default rate reaching a record 6.0% in April 2026.
- Distressed restructurings: Approximately 65% of all 2025 corporate defaults were distressed exchanges: not outright payment failures, meaning headline default statistics understate latent credit stress.
- Proskauer Private Credit Default Index (Q2 2026): Tracked 716 loans representing $195.6 billion in principal; recorded a 2.51% default rate for the quarter.
- First-lien senior secured recovery rates: Moody's current expected ultimate recovery on U.S. first-lien loans is approximately 68%: below the long-run historical average in the high-70s, due to the prevalence of loan-only capital structures with little junior debt cushion beneath them.
- Typical direct lending terms (2025–2026): First-lien senior secured loans at SOFR + 400–600 bps, maximum leverage covenants around 4x–5x EBITDA, and minimum interest coverage near 2.25x–2.5x.
- Nuveen Churchill Direct Lending portfolio (Q1 2026): Weighted average net leverage of 5.09x and interest coverage of 2.27x across first-lien loans, with approximately 86.86% of debt investments carrying financial maintenance covenants.
Sources: Moody's (April 2026); Fitch Ratings (May 2026); Proskauer Private Credit Default Index (Q2 2026); Nuveen Churchill Direct Lending Corp. Form 10-Q (Q1 2026).
Common Red Flags in EBITDA Normalization
- Widening gap between EBITDA and operating cash flow: A divergence between reported earnings and actual cash generation is the classic signal that adjustments are not converting to real cash flows.
- Recurring costs labeled non-recurring: IT upgrades, management fees, and restructuring charges that appear in multiple periods are not one-time items. Lenders should check all three years of financials before accepting the classification.
- Forward synergies in the LTM base: Projected cost savings that have not yet been implemented do not belong in trailing EBITDA. They belong in the buyer's upside case, not the lender's base case.
- Owner compensation added back above market rate: Only the premium over arm's-length market compensation is a legitimate add-back. The full salary is not.
- SBC in the add-back schedule: The SEC, FASB, and most institutional lenders treat stock-based compensation as a genuine economic cost. Underwriters who accept SBC add-backs are underwriting against a fictitious earnings base.
Underwriting Checklist: EBITDA Normalization Diligence
Use the following checklist on every deal before accepting any borrower-provided Adjusted EBITDA figure as the basis for leverage and coverage calculations:
Documentation Requirements
- ✅ Audited financials (3 years) or CPA-reviewed statements plus tax returns, verify tax return EBITDA aligns with CIM figures before proceeding
- ✅ Independent Quality of Earnings (QoE) report from a recognized accounting firm
- ✅ Full add-back schedule with line-item documentation for each adjustment
- ✅ Bank statements to reconcile reported cash flows (12 months minimum)
- ✅ Payroll records supporting executive compensation benchmarking
Analytical Questions to Ask
- ❓ Does the add-back appear in only one period, or has it recurred across multiple years?
- ❓ Is the gap between reported EBITDA and adjusted EBITDA greater than 20%? If so, each add-back must be re-verified independently.
- ❓ Are any synergies or cost-savings included that have not yet been realized? If yes, exclude from base case.
- ❓ Is stock-based compensation present in the add-back schedule? Reject unless the lender's credit policy explicitly permits it.
- ❓ What is the cash conversion ratio (operating cash flow / EBITDA)? Ratios consistently below 70% warrant deeper investigation into working capital dynamics.
- ❓ Does the FCCR remain above 1.20x under a 20% EBITDA stress scenario?
Metrics to Confirm Before Credit Approval
- ✅ Lender-underwritten Debt/EBITDA ≤ policy maximum (typically 4.0–5.0x for senior secured direct lending)
- ✅ Interest Coverage Ratio ≥ 2.0x on lender EBITDA
- ✅ Fixed Charge Coverage Ratio ≥ 1.20x post-CapEx, post-tax
- ✅ Stress scenario leverage and coverage remain within covenant headroom
- ✅ Covenant definitions in the credit agreement mirror the lender's underwritten EBITDA definition, not the seller's more expansive version
Key Takeaways
- The seller's Adjusted EBITDA is a negotiating position, not a credit fact. Every lender must independently rebuild the earnings figure before sizing leverage or setting covenants, seller-provided schedules are a starting point, not a conclusion.
- Stock-based compensation and unrealized synergies are the two most dangerous add-back categories. Both inflate leverage ratios, and neither converts to real cash that will service debt. Institutional lenders in 2026 have largely stopped accepting SBC add-backs regardless of QoE treatment.
- Distressed restructurings now dominate the default landscape. With approximately 65% of 2025 defaults classified as distressed exchanges rather than outright payment failures, headline default rates materially understate portfolio stress. Lenders who relied on reported metrics missed the signal.
- Recovery rates are declining structurally. Expected first-lien recoveries have drifted down to approximately 68% from historical highs above 77%, largely due to loan-only capital structures with no junior cushion. Every underwriting model should use 60–68% as the base-case first-lien recovery assumption, not historical averages.
- Covenant design must mirror underwriting discipline. The EBITDA definition in the credit agreement must reflect the lender's underwritten earnings base, not the borrower's expanded definition. A covenant built on inflated EBITDA provides no early warning and no protection.
