Underwriting & Risk Management, Bond Capital Editorial Desk
September 9, 2026
The Underwriting Challenge: When EBITDA Lies
A sponsor presents a manufacturing company with $12 million in “adjusted EBITDA.” The CIM is polished. The add-back schedule runs four pages. The headline leverage multiple looks tight. Then your QoE team opens the tax returns.
Three items flagged immediately: a $900,000 “one-time” restructuring charge that has appeared in three consecutive years; a $600,000 add-back for a departed executive whose replacement has already been hired at market rate; and $450,000 in projected cost synergies from a platform acquisition that closed eight weeks ago with no integration plan in place. Strip those three items and adjusted EBITDA falls to $10.05 million, a 16% reduction that blows the leverage covenant and reshapes the entire capital structure.
This scenario plays out across private credit markets with measurable regularity. EBITDA normalization is not a clerical exercise. It is the single most consequential analytical judgment a credit underwriter makes, because every multiple, every covenant, and every recovery assumption is anchored to it. Getting it wrong inflates debt capacity on paper while eroding it in practice, exactly the dynamic driving the credit stress visible across the middle market today.
The Normalization Framework: From Reported to Underwritable EBITDA
Experienced lenders treat the seller-provided adjustment schedule as a starting point, not a conclusion. The underwriting objective is to arrive at a “Lender-Underwritten EBITDA”, a defensible, conservative estimate of the sustainable earnings the business will generate under new ownership and under the proposed debt structure.
The Four-Step Normalization Process
- Anchor to Reported Financials. Begin with GAAP net income or operating income sourced from tax returns and audited statements, not management-prepared schedules. Tax returns are the single best anti-manipulation check because they carry legal consequence. Any material gap between the CIM’s stated net income and the tax return net income is an immediate diligence flag.
- Apply the Add-Back Test. For each proposed adjustment, apply three criteria: (a) Is it truly non-recurring? (b) Is it documented by invoices, agreements, or third-party evidence? (c) Would the cost actually disappear or change under the new ownership structure? Add-backs that fail any criterion are rejected or haircut.
- Apply Normalization Adjustments Downward. Normalization is bidirectional. If the company pays below-market rent to a related-party landlord, lender-underwritten EBITDA must be burdened with market rent. If the founder’s compensation is below what a professional CEO would cost, a market-rate replacement cost is added back as a deduction. A credible adjustment schedule includes both add-backs and add-downs, even when sellers resist presenting the latter.
- Cap Run-Rate and Pro Forma Items. Anticipated synergies and run-rate savings from future actions receive zero credit or a deeply haircut credit until demonstrated in actual financials. Post-close lender updates should track realized versus planned savings on a monthly basis.
Working Example: $10M EBITDA Manufacturing Company
Consider a sponsor-backed precision components manufacturer with the following bridge:
| Line Item | Sponsor-Presented ($000s) | Lender-Underwritten ($000s) | Underwriter’s Rationale |
|---|---|---|---|
| Reported EBITDA (GAAP) | $7,400 | $7,400 | Verified against tax return |
| (+) Restructuring Charges | $900 | $0 | Recurring in 3 of last 4 years, rejected |
| (+) Departed Executive Comp | $600 | $0 | Replacement hired at market rate, rejected |
| (+) Transaction Legal & Advisory Fees | $350 | $350 | Documented, deal-specific, accepted |
| (+) Non-Cash Stock-Based Compensation | $275 | $138 | 50% haircut applied; recurring economic cost recognized |
| (+) Projected Cost Synergies | $450 | $0 | No integration plan; post-close realization required |
| (−) Below-Market Related-Party Rent | $0 | ($185) | Market rate normalization applied downward |
| Lender-Underwritten EBITDA | $9,975 | $7,703 | −22.8% from sponsor presentation |
This 22.8% reduction directly affects debt capacity. At a 5.0x leverage ceiling, the sponsor’s EBITDA supports $49.9M in debt; the lender-underwritten EBITDA supports only $38.5M: an $11.4M structural gap that requires a price reduction, additional equity, or a deal that does not close.
Connecting EBITDA to Coverage and Leverage: The Credit Math That Matters
Once lender-underwritten EBITDA is established, it feeds directly into the two most critical credit metrics: the leverage ratio (Total Debt / EBITDA) and the debt service coverage ratio (DSCR).
Key Ratios and Their Thresholds
- Leverage Ratio (Total Debt / EBITDA): For smaller LBO transactions, lenders typically size debt at 3.0x to 4.0x EBITDA, with larger deals ranging between 4.0x and 5.5x. Using the manufacturer example above: $38.5M debt at 5.0x on the haircut EBITDA compares to an implicit 6.5x on sponsor-presented EBITDA, the exact kind of silent overleveraging that precedes covenant breach.
- DSCR (EBITDA − CapEx − Taxes) / Debt Service: Institutional private credit lenders generally require a minimum DSCR of 1.25x, with 1.50x or higher preferred for cyclical businesses. A DSCR of 1.0x implies zero margin of safety, any earnings softness triggers a default-adjacent cash shortfall.
- Fixed Charge Coverage Ratio (FCCR): Calculated as (EBITDA − CapEx − Cash Taxes) / (Interest + Scheduled Principal + Lease Payments), FCCR is the most conservative and operationally honest coverage metric. It forces underwriters to account for mandatory capital reinvestment, a line item sellers routinely exclude from their adjusted EBITDA presentations.
Stress Test: What Happens When EBITDA Misses by 20%?
Apply a 20% EBITDA downside scenario, consistent with a moderate cyclical correction in manufacturing, to both the sponsor and lender EBITDA bases:
- Sponsor EBITDA base ($9.975M) under 20% stress → $7.98M → Leverage climbs to 6.3x. Covenant breach at 6.5x threshold is two bad quarters away.
- Lender EBITDA base ($7.703M) under 20% stress → $6.16M → Debt sized to $38.5M now carries 6.25x leverage on stressed earnings. DSCR at 5% all-in interest cost falls to approximately 1.15x, below the 1.25x maintenance floor.
The stress test illustrates why rigorous normalization is not conservatism for its own sake. It is the mechanism by which a lender builds the covenant headroom needed to engage before a credit event, not after.
What Institutional Lenders Know: Market Data and Structural Realities
The Credit Performance Context
The private credit market has expanded dramatically, crossing $2.1 trillion in AUM globally, while simultaneously facing rising credit stress that validates the importance of conservative normalization at origination. Key data points from institutional sources:
📊 Private Credit Risk Data Box: 2024–2026
- Middle Market Default Rates (Proskauer Private Credit Default Index): The default rate for core middle market borrowers (EBITDA $25M–$50M) reached 3.0% in Q3 2024, up from 2.5% a year earlier. By Q1 2026, Proskauer’s index tracking 697 loans totaling $189.2 billion recorded an overall private credit default rate of 2.73%, up from 1.84% just two quarters prior.
- Distressed Restructuring Dominance (Moody’s): Distressed restructurings (debt exchanges and maturity extensions agreed under duress) accounted for approximately 65% of all 2025 private credit defaults, meaning headline payment default figures materially understate the scope of credit stress in the market.
- Recovery Rates (S&P Global Ratings): Based on S&P’s analysis of middle-market CLO portfolios, defaulted loans were exchanged at approximately 80 cents on the dollar on average, favorable compared to first-lien BSL recoveries in the low-to-mid 70s range historically.
- KBRA At-Risk Population: In KBRA’s Q4 2024 surveillance covering 1,903 unique middle market sponsored borrowers representing $922 billion in debt, approximately 5% of companies were identified as most at risk of default, characterized by fundamentally weak performance and limited financial flexibility.
- Covenant-Loose Structuring Risk (S&P Global): A leveraged buyout at 5.0x debt/EBITDA may carry a maximum leverage maintenance covenant of 8.25x, nearly 40% headroom, effectively allowing severe EBITDA deterioration before lender enforcement rights are triggered. S&P flagged that this dynamic mirrors the erosion of protection seen in covenant-lite syndicated loans.
- Add-Back Concentration Risk: Credit committees at institutional lenders now flag deals where more than 25% of EBITDA comprises run-rate savings and non-cash add-backs, treating that concentration as a credit risk indicator rather than acceptable financial presentation.
Common Red Flags in EBITDA Normalization
- The Recurring “One-Time” Charge. Restructuring, integration, or legal charges that appear in multiple periods are by definition recurring operating costs. Accepting them as add-backs inflates EBITDA and sets covenants off a false baseline.
- Unrealized Synergy Credits. Pro forma adjustments for cost saves or revenue synergies that depend on post-close actions and have not been demonstrated in actual results deserve zero underwriting credit until realized, with monthly lender reporting to track realization against plan.
- Aggressive Non-Cash Add-Backs. Stock-based compensation (SBC) is economically real even when non-cash. In sectors with structurally high SBC, presenting EBITDA with and without that item, and setting covenants off the more conservative figure, is institutional best practice.
- Missing Add-Downs. Below-market related-party economics (rent, intercompany services, loans) must be normalized upward as costs. Failing to include add-downs produces an EBITDA figure that cannot be sustained under arms-length operations.
- Wide Gap Between Seller and Lender EBITDA. When the reported-to-adjusted gap exceeds 20–30% of EBITDA, the adjustment schedule warrants heightened scrutiny. A gap of this magnitude typically signals that aggressive normalization is carrying deal valuation rather than documenting genuine non-recurring items.
EBITDA Normalization Underwriting Checklist
Apply this checklist at the Term Sheet stage and again at Final Commitment. Discrepancies between the two stages are themselves a diligence finding.
📋 Documentation Requirements
- ☐ Three years of tax returns, reconcile net income to CIM stated EBITDA line by line
- ☐ Audited or reviewed financial statements for the last two fiscal years
- ☐ Monthly management accounts for the trailing twelve months (TTM)
- ☐ Independent Quality of Earnings report from a recognized accounting firm for transactions above $25M in debt
- ☐ All related-party transaction schedules (rent, management fees, intercompany services)
- ☐ Payroll records and employment agreements for any owner/executive compensation add-backs
- ☐ Invoices or contracts supporting each non-recurring charge
- ☐ Integration plan and implementation timeline for any acquisition-related synergy add-backs
🔍 Analytical Questions to Answer Before Approval
- ☐ Does each add-back pass the three-part test: truly non-recurring, documented, and eliminable under new ownership?
- ☐ Are restructuring charges appearing in more than one period? If yes, reclassify as recurring operating cost.
- ☐ What is the total add-back as a percentage of reported EBITDA? Flag if above 25%.
- ☐ Have add-downs been applied for below-market related-party economics?
- ☐ At lender-underwritten EBITDA, does leverage remain within policy limits under a 15–25% stress scenario?
- ☐ Does the DSCR at lender EBITDA meet the 1.25x minimum on a stressed basis?
- ☐ Are maintenance covenants set within a tight enough band (10–20% headroom maximum) to provide meaningful early warning?
- ☐ Is a post-close monitoring protocol in place for monthly tracking of realized versus projected add-backs?
Key Takeaways
- EBITDA normalization is the load-bearing wall of private credit underwriting. Every leverage multiple, DSCR calculation, and covenant threshold is anchored to it. Errors compound through the entire credit structure.
- Reject add-backs that cannot pass the three-part test: truly non-recurring, independently documented, and genuinely eliminable under new ownership. Recurring “one-time” charges, undocumented synergies, and founder-departure compensation must be treated with heavy skepticism.
- Normalization must be bidirectional. Apply add-downs for below-market related-party economics. A schedule that includes only upward adjustments is a marketing document, not an underwriting document.
- Cap add-back concentration risk. When run-rate and non-cash add-backs exceed 25% of EBITDA, require independent validation, apply haircuts, and set covenants off the more conservative figure until realization is demonstrated in actual results.
- Stress test before commitment. A 20% EBITDA downside scenario applied to lender-underwritten EBITDA is the minimum standard. If a moderate stress breaks leverage covenants or DSCR floors, the deal is mispriced or over-levered, regardless of how the sponsor’s model presents it.
