EBITDA Adjustments and Normalization: The Underwriter’s Ground Truth in Private Credit

The Underwriting Problem Hidden in Plain Sight

Every private credit deal begins with a number, adjusted EBITDA, and in the current market, that number is under more scrutiny than at any point in the asset class’s modern history. When a sponsor presents a manufacturing business generating $10 million in “normalized” EBITDA and requests a $55 million unitranche facility, the underwriter’s first job is not to size the loan. It is to interrogate the denominator.

The stakes are material. According to the Financial Stability Board’s May 2026 Report on Vulnerabilities in Private Credit, debt-to-EBITDA ratios in private credit typically run at 5–6x on reported figures, but once EBITDA adjustments are stripped out, true leverage may approach 7x, a gap that represents a significant and systematic misrepresentation of borrower risk. A separate FSB-cited study of middle-market CLO borrowers found that 10% lacked sufficient cash flow to cover interest payments, with many carrying high debt-to-EBITDA ratios.

Meanwhile, Fitch Ratings reported a U.S. private-credit-backed corporate borrower default rate of 9.2% for 2025, and as of May 2026, the monthly rate had reached a record 6.0% per Fitch. The common thread across distressed credits is not macro surprise; it is over-leveraging driven by optimistic EBITDA normalization at origination. Getting the denominator right is the foundational discipline of sound credit underwriting.

The Normalization Framework: From Reported to Lender-Accepted EBITDA

EBITDA normalization is the process of adjusting a company’s reported earnings before interest, taxes, depreciation, and amortization to reflect true, recurring, and sustainable operating cash generation. The objective is not to maximize a headline figure; it is to construct a defensible baseline that withstands downside scenarios.

Experienced lenders work through a four-category taxonomy when evaluating a borrower’s add-back schedule:

1. Owner/Management Compensation Normalization: Adjusting above- or below-market compensation paid to owner-operators to a market-rate equivalent. This is typically the largest add-back category in lower middle market transactions and is generally accepted when documented with comp benchmarking data.
2. Genuinely Non-Recurring Items: One-time legal settlements, fire losses, one-time facility moves, or pandemic-era disruptions. These require documentary evidence (invoices, legal billings, board resolutions) and a five-year lookback to confirm they have not recurred.
3. Non-Cash Charges: Depreciation and amortization are already excluded from EBITDA by definition, but other non-cash items such as impairment charges, stock-based compensation (SBC), and unrealized losses may appear. SBC warrants particular caution: for high-SBC technology companies, adding back stock-based compensation without disclosure can inflate normalized EBITDA by 10–20%.
4. Run-Rate and Pro Forma Adjustments: Annualizing recently completed acquisitions, cost saves from headcount reductions, or pricing changes not yet fully reflected in trailing results. These are the most contentious category and carry the highest risk of overstatement.

Example, A $10M EBITDA Manufacturing Company: A sponsor presents a $10M adjusted EBITDA bridge for a precision components manufacturer. Reported EBITDA is $7.2M. Add-backs total $2.8M: $800K owner comp normalization, $400K one-time legal settlement, $600K SBC, $500K restructuring charges, and $500K pro forma savings from a plant consolidation completed three months prior. The lender’s credit committee accepts $1.6M in add-backs (owner comp and the legal settlement with documentation) and rejects $500K of the restructuring (recurred in three of the past five years) and applies a 50% haircut on the $500K pro forma savings pending a 12-month track record. Lender-accepted EBITDA: $8.8M versus the sponsor’s $10M: a 12% reduction that meaningfully changes leverage analysis at a 5.5x debt/EBITDA entry multiple.

Walking the Numbers: Leverage, Coverage, and Stress

Once a lender-accepted EBITDA figure is established, the core underwriting metrics can be calculated with integrity. The three most critical ratios are Total Debt/EBITDA (leverage), Fixed Charge Coverage Ratio (FCCR), and Free Cash Flow Conversion.

Leverage Analysis: Using the example above, if total debt is $48.4M against lender-accepted EBITDA of $8.8M, the entry leverage is 5.5x, a level that the Federal Reserve’s Interagency Guidance on Leveraged Lending flags as raising concerns for most industries when Total Debt/EBITDA exceeds 6x. Had the lender accepted the full $10M EBITDA, leverage would have appeared to be only 4.84x, a materially different risk profile that could pass internal screens without scrutiny.

Fixed Charge Coverage Ratio (FCCR): FCCR = (EBITDA − CapEx − Cash Taxes) ÷ (Cash Interest + Required Amortization). For our manufacturer with $1.2M capex, $800K cash taxes, $3.2M cash interest, and $1.0M amortization: FCCR = ($8.8M − $1.2M − $0.8M) ÷ ($3.2M + $1.0M) = $6.8M ÷ $4.2M = 1.62x. Most institutional lenders require a minimum FCCR of 1.10x to 1.25x at origination with a covenant set at 1.05x to 1.10x. A 1.62x FCCR provides meaningful cushion, but only when built on a defensible EBITDA base.

Stress Testing: Regulators and institutional lenders require at least one realistic downside scenario. A 15% revenue decline (consistent with a moderate industrial recession) flowing to a 25% EBITDA decline (operating leverage effect) produces stressed EBITDA of $6.6M, stressed leverage of 7.3x, and a stressed FCCR of 1.10x, right at covenant trigger. This confirms the deal has limited room for error and warrants tighter covenant architecture and conservative hold sizing.

What Institutional Lenders See, and What Concerns Them

The market-wide data on EBITDA normalization in 2025–2026 provides important context for underwriters evaluating add-back schedules. Proskauer’s 15th Annual Private Credit Insights Report found that 57% of European private credit deals now include revenue adjustments as part of EBITDA normalization, reflecting a broadening of what sponsors seek to add back. The same report noted that independent due diligence certifications on those adjustments are declining, a troubling combination.

The FSB’s May 2026 report highlights that approximately 75% of private credit borrowers have EBITDA below $100 million, and that these smaller issuers tend to have less transparent financial reporting, making EBITDA normalization disputes harder to resolve in post-close monitoring. Moody’s also flags that payment-in-kind income as a share of BDC investment income continued to rise through early 2026, a signal that some borrowers are effectively borrowing against future EBITDA to service current obligations.

The divergence between the CDLI-S’s 0.29% loss rate and Fitch’s 9.2% default count illustrates a critical distinction: the Cliffwater index measures realized and unrealized credit losses as a percentage of assets, while Fitch counts each default event regardless of recovery. Well-underwritten senior secured positions with accurate EBITDA baselines can experience default events while suffering minimal economic loss, but that outcome depends entirely on whether lenders sized correctly against true cash flow from the start.

The EBITDA Normalization Underwriting Checklist

The following diligence framework should be applied to every add-back schedule before a credit is approved:

1. Source Every Add-Back to a Primary Document
   Each item must be traceable to a specific P&L line, an invoice, a legal billing, or a payroll record. If an adjustment cannot be mapped to the accounting records, it should not be assumed valid. Reject anything that cannot be sourced.
2. Conduct a Five-Year Frequency Analysis
   Pull five years of financial statements and code each “non-recurring” item by year. Companies that restructure annually or face frequent litigation have recurring costs masquerading as one-time items. Any item appearing in more than two of the past five years should be treated as recurring.
3. Apply Conservative Treatment to Pro Forma Savings
   Cost saves from plant consolidations, headcount reductions, or vendor renegotiations should not receive full credit unless the saving has been realized for at least two full fiscal quarters. Apply a 50% haircut on unproven savings and require a clawback mechanism in the credit agreement if savings fail to materialize.
4. Scrutinize Stock-Based Compensation Add-Backs
   SBC represents real economic dilution to equity holders and real compensation cost to the business. Many institutional lenders exclude SBC when calculating debt capacity, because lenders underwriting leverage ratios should not extend credit against earnings inflated by adding back genuine compensation costs.
5. Reject Synergy and Forward-Looking Adjustments
   Future growth and buyer-side synergies belong in the buyer’s own underwriting model, not in the seller’s add-back schedule. Normalized EBITDA describes historical, repeatable cash, not aspirational projections.
6. Commission an Independent Quality of Earnings (QoE) Report
   For transactions above $5M EBITDA, require a third-party QoE from an independent accounting firm. The QoE should validate the add-back schedule, assess revenue quality, and stress-test working capital assumptions. The lender’s credit team should review the QoE independently, not simply accept the sponsor’s commissioned version.
7. Calculate LTM EBITDA with Precision
   LTM construction requires matching stub periods to the day and applying consistent accounting policies across both stubs. The stub-period adjustment is the most error-prone step in LTM construction, a six-month current stub demands a six-month prior stub on the same accounting basis.
8. Run a Downside Scenario at Lender-Accepted EBITDA
   Per Federal Reserve guidance, projections should include one or more realistic downside scenarios that reflect key risks identified in the transaction. Model a 15–25% EBITDA haircut and confirm that FCCR remains above 1.0x and leverage remains serviceable without covenant breach.

Key Takeaways

• The denominator is the deal. A 10–15% reduction in lender-accepted EBITDA versus sponsor-presented EBITDA can shift a credit from investment-grade quality to a covenant-sensitive position at the same debt quantum. Every underwriting decision flows from this number.
• EBITDA adjustments systematically mask true leverage. The FSB’s May 2026 analysis found that reported 5–6x leverage in private credit may reflect true leverage closer to 7x once add-backs are stripped out, well above the Federal Reserve’s 6x concern threshold for most industries.
• Five-year frequency analysis is non-negotiable. “Non-recurring” items that recur are, by definition, recurring costs. A structured lookback is the only reliable way to distinguish genuine one-time events from chronic operational leakage that erodes true cash generation.
• Default data diverges sharply by methodology. The gap between Cliffwater’s 0.29% senior loss rate and Fitch’s 9.2% borrower default count is not a contradiction; it reflects recovery performance. Well-secured, properly sized positions survive default events with minimal loss. The underwriting discipline that enables that outcome starts with an accurate EBITDA base.
• Stress testing must precede approval, not follow it. Regulators require realistic downside scenarios at origination. Lenders who model stress at lender-accepted EBITDA, not sponsor-presented EBITDA, identify covenant vulnerabilities before they become portfolio problems.