The Underwriting Challenge: When the Numbers Don't Tell the Whole Story

You are reviewing a Confidential Information Memorandum for a $10 million EBITDA manufacturing company seeking a leveraged buyout facility. The sponsor's EBITDA bridge starts at $7.2 million reported and lands at $10.0 million after add-backs: a one-time restructuring charge, management bonuses “above run-rate,” a CEO replacement cost, stock-based compensation, and a grab-bag of “synergies to be realized.” The proposed leverage is 5.0x. On paper, the deal looks fundable. In reality, you may be underwriting air.

This is the central tension in private credit underwriting. EBITDA normalization is a legitimate and necessary process — no lender should penalize a borrower for a genuinely non-recurring legal settlement or a one-time facilities overhaul. But the bridge between reported and adjusted EBITDA has become one of the highest-risk areas in deal assessment. As one advisory firm noted, that bridge is now “a red-flag in the eyes of investors and diligence teams that needs to be proactively addressed.”

The stakes are structural. Because most middle-market credit agreements set leverage covenants, call protection triggers, and pricing grids directly against a defined EBITDA figure, every dollar of unjustified add-back inflates the reported cushion, compresses perceived leverage, and allows a borrower to incur more debt than their operating cash flow can sustainably support. For lenders, the consequence is not abstract: it is a higher probability of covenant breach, credit deterioration, and eventual loss.

This article provides a practitioner's methodology for dissecting EBITDA add-backs, constructing a defensible normalized earnings figure, stress-testing it against realistic scenarios, and identifying the red flags that experienced credit committees flag before committing capital.

The Framework: From Reported EBITDA to Underwriting EBITDA

A disciplined EBITDA normalization process moves through four sequential layers. Each layer applies increasing scrutiny and a different analytical lens:

Layer 1 — Reported EBITDA (Starting Point)

Begin with audited financials or, where unavailable, reviewed statements. Reconcile to tax returns. If the CIM's net income does not match the tax return, treat all add-backs as provisional until verified. This alignment check is non-negotiable; a mismatch is one of the most common early signals of inflated seller financials.

Layer 2 — Adjusted EBITDA (The Seller's Claim)

Adjusted EBITDA adds back non-recurring, non-cash, and owner-specific costs to reported EBITDA. The categories most frequently encountered in middle-market CIMs include:

• Non-recurring expenses: One-time litigation settlements, facility closures, regulatory remediation costs, or ERP implementation fees. Requires documentation that the event is genuinely non-repeating.
• Owner compensation normalization: Adjusting above-market or below-market owner salaries to a third-party market equivalent. Requires compensation benchmarks (e.g., BLS OEWS data, Robert Half industry guides).
• Related-party transactions: Below-market rent paid to an owner-affiliated entity must be restated to market rent; above-market management fees paid to a PE sponsor must similarly be normalized. As one practitioner framework notes, lenders scrutinize discretionary expenses and verify that “related-party transactions — such as below-market rent — are adjusted to reflect actual economic” conditions.
• Non-cash charges: Stock-based compensation (SBC), amortization of deal-related intangibles, and non-cash impairments. SBC in particular has become a flashpoint: institutional lenders and the SEC treat SBC as a real economic cost, and buyers underwriting against SBC add-backs have in practice reduced their accepted earnings multiple by 0.5 to 1.0 turns.
• Pro forma run-rate savings: Cost eliminations expected post-close (e.g., redundant headcount, facility consolidation). The 2025–2026 market standard has moved to permitting pro forma synergies in covenant calculations only when they are realized within 18 months and validated by an independent accounting firm.

Layer 3 — Normalized EBITDA (The Lender's Independent View)

Normalized EBITDA is the lender's own independently derived figure — not the seller's schedule. It reflects sustainable free cash flow under realistic steady-state operating conditions, including market-rate staffing, maintenance capital expenditure, and required reinvestment. This figure will almost always be lower than the seller's adjusted EBITDA. The delta between the two is the “EBITDA credibility gap” — and sizing that gap is the core underwriting task.

Layer 4 — Underwriting EBITDA (Stressed Base Case)

The underwriting EBITDA applies a conservatism haircut to normalized EBITDA to account for execution risk on unrealized add-backs, cyclical revenue sensitivity, and margin pressure from input cost inflation or customer concentration. This is the figure used to size debt, set initial leverage, and establish maintenance covenants.

Key Principle: Seller-provided normalization schedules are a starting point, not a conclusion. Lenders must independently assess every adjustment to determine whether it is reasonable, documented, and supported by third-party evidence.

The Analysis: Walking Through a $10M EBITDA Manufacturing Company

Consider the following EBITDA bridge for a hypothetical mid-market industrial distributor (illustrative, not drawn from any single transaction):

The delta is $875,000 — seemingly modest at 8.75% of adjusted EBITDA. But at a 5.0x senior leverage multiple, the sponsor's debt sizing assumption is $50.0 million. On the lender's normalized EBITDA of $9.125 million, that same debt load implies 5.48x leverage — a materially different risk profile, likely outside institutional leverage guardrails and certainly requiring a reset of covenant levels.

Applying the Debt Service Coverage Test

The Debt Service Coverage Ratio (DSCR) is calculated as:

DSCR = Normalized EBITDA − Cash Taxes − CapEx ÷ (Annual Principal + Interest)

Assume: $9.125M normalized EBITDA, $600K cash taxes, $900K maintenance CapEx = $7.625M free cash flow. On $50M of debt at 8.5% all-in cost with 1% annual amortization = $4.75M annual debt service.

DSCR = $7.625M ÷ $4.75M = 1.60x — acceptable, but thin. If the lender had accepted the seller's $10M EBITDA without challenge, the apparent DSCR would be 1.79x, creating a false sense of headroom. Stress it down 15% in a mild revenue contraction and DSCR falls to 1.36x — approaching institutional minimum thresholds typically set at 1.25x.

Industry Standards: What Institutional Lenders Require in 2026

The credit market's tolerance for aggressive EBITDA normalization has compressed materially since the low-rate era of 2020–2021. Several structural forces are driving tighter underwriting discipline:

1. Covenant tightening in response to elevated defaults. According to Moody's July 2025 Credit Strategy report, leveraged loan default rates were projected to reach 7.5% by year-end 2025 and peak at 7.9% in Q1 2026 — nearly double the 3.4% historical average. The Federal Reserve's 2025 Senior Loan Officer Survey confirmed that senior debt covenants tightened through 2025 as the credit cycle matured, forcing sponsors to underwrite tighter Adjusted EBITDA definitions. Against this backdrop, inflated add-back schedules carry direct portfolio consequences.

2. The SBC add-back reversal. In 2021, stock-based compensation add-backs were standard practice in PE-backed middle-market credit agreements. By 2026, sponsor LOIs in the lower middle market routinely exclude all SBC add-backs. This reflects a growing consensus that SBC represents a genuine compensation cost that will recur in any management-intensive business.

3. Synergy discipline with time gates. The market has adopted a formal realization standard: pro forma synergies may be included in covenant EBITDA only if they are documented, third-party validated, and realized within 18 months of closing. Any synergy claim exceeding this window should receive a significant haircut at underwriting.

4. Add-back concentration risk. Credit committees now flag transactions where more than 25% of Adjusted EBITDA is derived from run-rate savings and non-cash add-backs, treating that concentration as a structural credit risk rather than legitimate financial engineering. If independent validation is unavailable, lenders should haircut underwriting EBITDA for covenant-setting purposes.

5. Mid-market private credit loss context. For calibration, the Cliffwater Direct Lending Index (CDLI) — which tracks approximately 21,000 loans and $549 billion in assets — reported a 20-year average annual credit loss rate of 1.01%, with Q2 2025 losses tracking at just 0.75% annually. The CDLI-Senior specifically reflects conservative senior secured structures. Mid-market private credit has historically delivered lower loss rates relative to broadly syndicated leveraged loans due to tighter structures and stronger lender protections. That differential is not accidental: it is the direct product of rigorous EBITDA underwriting and covenant discipline.

Underwriting Checklist: EBITDA Normalization Diligence

The following checklist reflects institutional best practices for independently verifying and stress-testing EBITDA add-back schedules in middle-market credit underwriting:

1. ✅ Reconcile reported EBITDA to tax returns and audited financials.
   Request three years of tax returns and two years of audited (or reviewed) statements. Any discrepancy between the CIM's P&L and the tax return triggers a full line-item reconciliation before add-backs are considered.
2. ✅ Categorize every add-back by type.
   Separate into: (a) non-recurring cash expenses, (b) non-cash charges, (c) owner compensation adjustments, (d) related-party normalizations, and (e) pro forma run-rate savings. Each category requires different documentation and scrutiny.
3. ✅ Require primary-source documentation for every add-back above $50,000.
   Invoices, board resolutions, legal settlement agreements, payroll records, and market lease comps. Verbal representations or CIM summaries are insufficient.
4. ✅ Apply a “would this cost disappear under new ownership?” test.
   The standard is traceability and reasonableness: Can the adjustment be traced to records? Is the rationale clear? Would the cost truly change under new ownership? If the answer to any of these is uncertain, apply a haircut.
5. ✅ Benchmark owner compensation to third-party market data.
   Use BLS Occupational Employment and Wage Statistics or industry-specific compensation guides. Do not accept self-reported “market rate” claims without independent verification.
6. ✅ Reject or heavily haircut SBC add-backs.
   Treat stock-based compensation as a recurring economic cost in your underwriting model, consistent with SEC, FASB, and institutional lender standards. If retained in the credit agreement EBITDA definition, require a dual EBITDA presentation for internal credit committee review.
7. ✅ Apply an 18-month realization gate to all run-rate synergies.
   Cap pro forma savings at a percentage of normalized EBITDA (typically 10–15%) and require independent accounting firm validation for any synergy above that cap. Synergies expected to materialize beyond 18 months post-close should be excluded from underwriting EBITDA.
8. ✅ Flag transactions where add-backs exceed 25% of Adjusted EBITDA.
   High add-back concentration is a structural credit risk indicator. Require independent Quality of Earnings (QoE) analysis commissioned by the lender (not the borrower or sponsor) for any transaction meeting this threshold.
9. ✅ Calculate DSCR and leverage on the lender's normalized figure, not the seller's.
   Model DSCR at 1.0x, 1.25x, and 1.5x the expected annual debt service to establish entry conditions and early warning levels. Stress EBITDA by 15–20% for sensitivity analysis.
10. ✅ Document the EBITDA gap and present it to credit committee.
    The delta between the seller's Adjusted EBITDA and the lender's Normalized EBITDA must be formally documented, with each line-item variance explained and the impact on leverage and coverage explicitly calculated.

Key Takeaways

• EBITDA normalization is a lender discipline, not a borrower courtesy. Seller-provided add-back schedules are a starting point. Every adjustment requires independent documentation, verification, and economic rationale before a lender credits it in their underwriting model.
• The EBITDA gap directly determines capital structure risk. A seemingly modest 8–10% inflation in Adjusted EBITDA translates into meaningfully higher leverage multiples at closing and materially thinner DSCR headroom — two of the most predictive indicators of future credit stress.
• SBC and unvalidated synergies are the highest-risk add-back categories in 2026. Market practice has shifted decisively: stock-based compensation is now widely retained as an expense in institutional lender underwriting models, and pro forma synergies require 18-month realization gates and third-party validation.
• Add-back concentration is itself a risk signal. When more than 25% of a borrower's Adjusted EBITDA is derived from non-cash and run-rate add-backs, credit committees should require a lender-commissioned QoE analysis and consider a floor haircut on underwriting EBITDA for covenant purposes.
• Default data validates underwriting rigor. With leveraged loan default rates tracking nearly double their historical average and distressed exchanges comprising the majority of credit events, the cost of relaxed EBITDA standards is not theoretical. Mid-market private credit's historically lower loss rates — as measured by the Cliffwater CDLI — are directly attributable to tighter covenant structures and more conservative earnings underwriting.

Sources & References

1. Moody's Ratings — US Credit Review & Outlook: Corporate Credit Risk, July 2025. moodys.com
2. Cliffwater LLC — Cliffwater Direct Lending Index (CDLI). Referenced via Morgan Stanley Investment Management, 2025. morganstanley.com
3. PKF O'Connor Davies — Bridging the EBITDA Credibility Gap: A Guide for Sellers, September 2025. pkfod.com
4. ClearlyAcquired — Understanding Financial Normalization in M&A, 2026. clearlyacquired.com
5. Ledgerism — EBITDA Adjustments 2026: Real Examples & Buyer Pushback, June 2026. ledgerism.net
6. Ferdinand Partners — Normalized EBITDA Adjustments: What Sellers Add Back and Why Buyers Should Challenge. ferdinandpartners.com
7. Private Equity Bro — EBITDA Add-Backs in Private Credit: Rules, Risks & Realities, October 2025. privateequitybro.com
8. Pioneer Capital Advisory — EBITDA Normalization Explained for Buyers. pioneercapitaladvisory.com
9. Angel Investors Network — Private Credit Default Rates 2025: What the Data Shows, May 2026. angelinvestorsnetwork.com
10. CoinLaw — Corporate Bond Default Statistics 2026. coinlaw.io
11. CT Acquisitions — LTM EBITDA in 2026: The Trailing Twelve Months Metric Explained, August 2026. ctacquisitions.com
12. Northleaf Capital Partners — Private Credit Market Update: Q3-2025, November 2025. northleafcapital.com