July 13, 2026 | Underwriting & Risk Management
EBITDA Adjustments & Normalization: The Credit Underwriter’s Framework for Separating Real Earnings from Financial Noise
When a sponsor presents $12 million in adjusted EBITDA to support a $60 million senior credit facility, the underwriter’s first question is not what is the leverage ratio, it is which version of EBITDA are we actually multiplying? That distinction is the entire ballgame.
The Underwriting Challenge
EBITDA normalization sits at the intersection of accounting judgment, negotiation strategy, and genuine credit risk. In leveraged finance and private credit transactions, the adjusted EBITDA figure presented in a Confidential Information Memorandum (CIM) or Quality of Earnings (QoE) report directly determines how much debt a company can carry, what covenants are set at, and whether the deal pencils at all.
The stakes are asymmetric. On the upside, a well-documented add-back schedule improves credit availability and pricing for a borrower. On the downside, an inflated EBITDA number that survives initial underwriting but collapses post-close produces an over-levered borrower, a covenant breach within twelve months, and a workout that destroys lender value.
The market environment makes this discipline more critical than ever. Moody’s data shows that distressed exchanges, often triggered by over-leveraged capital structures built on aggressive EBITDA underwriting, accounted for 64% of all defaults in the first half of 2025, approximately the same pace as all of 2024, which itself was the highest annual share on record. Understanding exactly what earnings a business can sustainably generate is not a valuation exercise. It is a loss prevention discipline.
This article walks through the methodology institutional credit underwriters use to assess, challenge, and stress-test EBITDA normalization, from the initial add-back audit to a fully modeled downside scenario.
The Analytical Framework: From Reported to Underwritten EBITDA
Experienced lenders distinguish between three versions of EBITDA that any borrower presents, and they are not the same number:
- Reported EBITDA: Taken directly from audited or reviewed financial statements. No adjustments. The floor of the analysis.
- Adjusted EBITDA: Management’s bridge from reported results to a normalized earnings figure, incorporating add-backs for non-recurring or non-operating items. This is the number the borrower presents. It is a starting point, not a conclusion.
- Normalized (Underwritten) EBITDA: The earnings figure the lender independently validates through diligence. It reflects adjustments that survive scrutiny: traceable to source documents, non-recurring by evidence, and not masking an underlying cost the new capital structure must absorb.
The gap between adjusted and normalized EBITDA is where credit risk lives. Sellers and borrowers often focus on the highest possible adjustment case. Buyers and lenders focus on which earnings will continue after closing, which costs must be replaced, and which assumptions belong in upside scenarios rather than base-case underwriting. The practical issue is not whether a company can identify possible adjustments; it is whether the adjustments represent recurring economic value that can support a debt obligation.
Lenders are structurally cautious. They use the most conservative EBITDA definition because it drives debt sizing, covenant headroom, and repayment capacity. An adjustment that cannot be clearly justified and traced to the company’s financial statements will not be counted when determining how much financing a borrower can access.
Example: Analyzing a $10M Reported EBITDA Manufacturing Company
A mid-market industrial manufacturer reports $10.0M in trailing twelve-month EBITDA. The sponsor’s CIM presents an adjusted EBITDA of $12.4M, a 24% uplift, to support a proposed $62M first-lien term loan at 5.0x leverage.
| Adjustment Item | Sponsor Claim | Lender Assessment | Accepted Amount |
|---|---|---|---|
| Excess owner compensation | $800K | Market replacement salary benchmarked at $350K; accept partial | $450K |
| One-time litigation settlement | $600K | Documented via court filing; confirmed non-recurring | $600K |
| “Restructuring” costs (Year 3 of 3) | $500K | Recurring pattern in three consecutive years, rejected | $0 |
| Management consulting fees (related party) | $400K | Contract not terminated post-close; cost will continue, rejected | $0 |
| Pro forma cost savings (not yet realized) | $100K | Partial credit pending implementation evidence | $50K |
| Total Add-Backs | $2,400K | $1,100K | |
| Normalized EBITDA (Underwritten) | $12.4M (Sponsor) | $11.1M (Lender) |
At a 5.0x leverage multiple, that $1.3M normalization gap reduces supportable debt from $62M to $55.5M, a $6.5M difference in credit exposure before a single covenant is written.
Walking the Calculations: From Normalized EBITDA to Credit Decision
Once the underwritten EBITDA is established, the lender runs three core calculations that govern credit structuring:
1. Total Leverage Ratio
Formula: Total Debt ÷ Normalized EBITDA
Using the example above: $55.5M debt ÷ $11.1M EBITDA = 5.0x leverage. The lender’s covenant is typically set at 5.5x or 6.0x with a 10–15% headroom buffer. If the sponsor had underwritten to the $12.4M figure, the lender would have been unknowingly operating at 5.6x on day one.
2. Interest Coverage Ratio (ICR)
Formula: Normalized EBITDA ÷ Annual Interest Expense
At SOFR + 450 bps with SOFR at ~4.3%, the all-in rate is approximately 8.75%. On $55.5M of debt, annual interest is ~$4.9M. ICR = $11.1M ÷ $4.9M = 2.27x. A minimum ICR covenant of 2.0x provides only 13% downside buffer, thin for a cyclical manufacturer. Lenders underwriting to the inflated $12.4M EBITDA would have calculated a false ICR of 2.53x, creating the illusion of additional cushion that does not exist.
3. Debt Service Coverage Ratio (DSCR)
Formula: (Normalized EBITDA − Taxes − CapEx − Working Capital Needs) ÷ Annual Debt Service
Free cash flow conversion is the critical variable most sellers understate. Adjusted EBITDA to free cash flow conversion must account for maintenance CapEx (often 3–5% of revenue in manufacturing), normalized working capital swings, and cash taxes. A business reporting $11.1M in normalized EBITDA may generate only $6.5–$7.5M in true free cash flow, a conversion ratio of 60–68%. If aggressive EBITDA normalization overstates earnings by 15–20%, DSCR may appear acceptable on paper but fail during underwriting when properly stress-tested.
Stress Testing: The -20% EBITDA Scenario
Every institutional credit underwriter models a downside case. A standard approach applies a 20% EBITDA stress to reflect a moderate revenue decline combined with cost inflexibility:
- Stressed EBITDA: $11.1M × 0.80 = $8.9M
- Stressed Leverage: $55.5M ÷ $8.9M = 6.2x (breaches a 6.0x covenant)
- Stressed ICR: $8.9M ÷ $4.9M = 1.82x (below a 2.0x floor)
This stress test is not a prediction; it is a structural question: does the credit survive a moderate adverse scenario without triggering a technical default that forces a costly amendment or accelerated restructuring? In this example, the answer is no. That finding either re-prices the deal, reduces permitted leverage, or adds structural protections before closing.
What Institutional Lenders Look For: Standards, Data, and Red Flags
Market Context: Default and Recovery Rates (2024–2025)
The consequences of poor EBITDA underwriting are not theoretical. Credit performance data from major rating agencies frames the risk environment within which normalization discipline operates:
📊 Key Credit Performance Data Points
- Leveraged Loan Default Rate (Moody’s, mid-2025): The leveraged loan default rate is projected to end 2025 at approximately 7.5%, nearly double the 3.4% long-run historical average, with loan-only capital structures backed by private equity firms exhibiting the most distress.
- Distressed Exchanges Dominate: Distressed exchanges accounted for roughly 64% of all defaults in the first half of 2025, the same pace as 2024, which itself was the highest annual share on record. Many of these reflect borrowers whose original EBITDA underwriting proved too optimistic.
- High-Yield Default Rate: High-yield bond issuers are projected at a 3.2% realized default rate for 2025, rising above 4% by early 2026, though still below the 4.5% historical average since 1996.
- First-Lien Recovery Rates (Moody’s): Moody’s estimates expected ultimate recoveries on U.S. first-lien loans at roughly 68%, down from long-run historical experience in the high 70s, driven by loan-only capital structures and liability management exercises (LMEs) that can strip collateral from non-participating lenders.
- Senior Secured vs. Unsecured Recovery: According to Moody’s recovery data across credit cycles, senior secured debt historically recovers 70–80 cents on the dollar, versus 40–50 cents for senior unsecured debt and 20–30 cents for subordinated unsecured obligations, underscoring why lien position and collateral quality remain the primary recovery determinants.
- Cov-Lite Loans: By 2021, covenant-lite loans represented more than 86% of total outstanding leveraged loan volume and more than 90% of new issuance. In the broadly syndicated market, maintenance covenants, the primary early warning mechanism for lenders, have become the exception rather than the rule.
Red Flags That Experienced Underwriters Cannot Ignore
Practitioners who have worked through credit cycles have learned to identify specific patterns in add-back schedules that signal structural earnings weakness. The following are the most consequential:
- Recurring “Non-Recurring” Costs: Add-backs that appear in multiple years of financial history are not non-recurring. Restructuring charges, integration costs, or “one-time” consulting fees that appear for three consecutive years are operating costs by another name.
- Aggressive Revenue Recognition: Unusual revenue timing, channel stuffing near quarter-end, or contract accounting that accelerates recognition ahead of cash collection inflates EBITDA without creating sustainable cash flow.
- Customer Concentration: Revenue durability becomes harder to underwrite when a small number of customers represent a large share of revenue. A borrower where 40%+ of revenue comes from a single customer requires scenario analysis for contract non-renewal.
- Deferred Maintenance CapEx: Sellers who reduce capital expenditure in the 18–24 months pre-sale to inflate EBITDA transfer that cost to the new ownership structure. Lenders must independently assess replacement CapEx needs.
- Working Capital Manipulation: Unusual swings in receivables, payables, or inventory near transaction close (accelerating collections or stretching payables) can temporarily inflate reported cash flows while degrading operational health.
- The Winner’s Curse in Competitive Auctions: Auction processes create structural pressure to accept aggressive add-backs. A buyer or lender that challenges add-backs risks losing the deal to a competitor with a more generous view of the seller’s adjustments. The winner is often the party with the most optimistic EBITDA underwriting, which also means the most concentrated risk of post-close underperformance.
The Underwriting Checklist: Step-by-Step Diligence Protocol
The following framework represents the minimum diligence standard for EBITDA normalization in a private credit or leveraged finance transaction:
✅ EBITDA Normalization Diligence Checklist
Phase 1: Document Collection
- ☐ Three years of audited or reviewed financial statements (P&L, balance sheet, cash flow)
- ☐ Three years of federal tax returns (verify CIM net income ties to tax return line items)
- ☐ Monthly management accounts for the trailing 24 months
- ☐ Payroll records and executive compensation detail
- ☐ Seller-provided add-back schedule with supporting documentation for each line item
- ☐ Third-party Quality of Earnings (QoE) report from a recognized accounting firm (if available)
Phase 2: Add-Back Audit
- ☐ Trace each add-back to source documents (invoices, contracts, court filings, payroll records)
- ☐ Test for recurrence: does the item appear in prior-year financials?
- ☐ Assess owner compensation against market replacement cost benchmarks
- ☐ Identify all related-party transactions (rent, management fees, intercompany services) and normalize to arm’s-length rates
- ☐ Evaluate whether pro forma cost savings have been implemented or remain aspirational
- ☐ Identify and quantify any “add-downs”, non-recurring income items that inflated reported EBITDA
Phase 3: Cash Flow Bridge
- ☐ Reconcile normalized EBITDA to free cash flow (subtract taxes, maintenance CapEx, required working capital investment)
- ☐ Assess historical CapEx relative to reported depreciation, a persistent CapEx-to-D&A ratio below 1.0x may signal underinvestment
- ☐ Analyze working capital trends: DSO, DPO, inventory turns over the trailing eight quarters
- ☐ Calculate DSCR under base case and stress scenario (minimum -20% EBITDA)
Phase 4: Covenant Calibration
- ☐ Set maximum leverage covenant with minimum 15–20% headroom to underwritten EBITDA (not sponsor-presented EBITDA)
- ☐ Set minimum ICR covenant at 1.75x–2.0x with quarterly testing requirement
- ☐ Cap total EBITDA add-backs in the credit agreement (a 40% add-back cap on non-adjusted EBITDA is an emerging institutional standard for sub-$50M EBITDA borrowers)
- ☐ Document the EBITDA definition in the credit agreement with specificity, vague definitions create amendment risk
Phase 5: Ongoing Monitoring
- ☐ Require quarterly compliance certificates with EBITDA bridge reconciliation
- ☐ Establish triggers for management call or site visit (leverage ratio >0.5x above covenant, revenue decline >10% year-over-year)
- ☐ Monitor EBITDA-to-cash conversion ratio each quarter; persistent deterioration signals accounting or operational stress
Key Takeaways for Credit Underwriters
- 🎯 The version of EBITDA matters as much as the multiple. In a deal priced at 5.0x, a $1M discrepancy in normalized EBITDA translates to $5M in over-leveraging. Normalization accuracy is leverage accuracy.
- 🚨 Recurring non-recurring costs are the most common fatal flaw. Any add-back that appears in two or more years of financial history fails the non-recurring test and must be rejected unless the underlying condition has been contractually resolved.
- 📉 Recovery rates are trending lower in a challenging normalization environment. With Moody’s estimating first-lien recovery rates near 68% (down from historical highs of 70–80%), and distressed exchanges stripping collateral from non-participating lenders, underwriting discipline at origination is the most reliable loss prevention mechanism available.
- 📋 Covenant design must be grounded in underwritten EBITDA, not sponsor-presented EBITDA. Covenants set with reference to an inflated earnings figure provide false comfort. Private credit lenders with maintenance covenant protections retain early warning mechanisms that broadly syndicated, cov-lite lenders have surrendered.
- 🔄 EBITDA-to-free-cash-flow conversion is the stress test that matters most. A 60–70% conversion ratio is typical for asset-intensive middle-market businesses. Underwriters who model the full cash waterfall (taxes, CapEx, working capital, debt service) identify structural repayment risk that headline EBITDA multiples systematically obscure.
