Adjusted EBITDA
The Battle of Adjustments
EV = Multiple × EBITDA. A small tweak to the denominator, amplified by the multiplier, shifts deal prices by hundreds of millions. That is why "one-time or recurring?" is the most contested question in every M&A negotiation — and why Adjusted EBITDA sits at the center of every deal structure, debt package, and incentive plan.
The Multiplication Effect
Enterprise Value = Multiple × EBITDA
This single formula explains why every M&A deal becomes a negotiation over EBITDA definitions. A $10M add-back, multiplied by a 15× deal multiple, is a $150M swing in purchase price.
Valuation impact of EBITDA add-backs by deal multiple ($ millions)
| Add-back ($M) | 10× | 12× | 15× | 18× | 20× |
|---|---|---|---|---|---|
| $5M | $50M | $60M | $75M | $90M | $100M |
| $10M | $100M | $120M | $150M | $180M | $200M |
| $25M | $250M | $300M | $375M | $450M | $500M |
At a 15× multiple, every $1M of EBITDA add-back translates to $15M of additional deal value.
Conflict-of-Interest Map
Every party involved in a leveraged deal has a financial incentive to tolerate — or actively inflate — the EBITDA figure. Understanding who benefits from what is the first step in reading any adjusted EBITDA disclosure critically.
Sell-side Advisory IB
M&A Advisory
Incentive: Maximize EBITDA
Success fees = 0.5–2% of deal value. A $10M add-back at a 15× multiple raises deal size by $150M → adds $1.5M in fees.
Management Team
MIP / Equity Rollover
Incentive: Hit EBITDA targets
Management Incentive Plans (MIPs) are tied to exit EBITDA or deal multiple. Management controls the data room while being the primary beneficiary of higher add-backs — the most acute conflict of interest.
Leveraged Finance Team
LevFin — Debt Origination
Incentive: Maximize loan size
LevFin fees = 1–2% of loan principal — bigger loan, bigger fee. Higher EBITDA allows more leverage (typical 5× EBITDA limit). After syndication, risk is off-balance-sheet — the core 'Originate to Distribute' structural problem.
Syndication / CLOs
Institutional Loan Investors
Incentive: Secure yield
Investment decisions based on credit memos written by LevFin banks. Covenant-lite structures remove EBITDA-linked covenants — no early warning when performance deteriorates. A key driver of the leveraged loan market expanding from $600B to $1.3T between 2015–2019.
Credit Rating Agencies
S&P / Moody's / Fitch
Incentive: Apply internal EBITDA standards
Issuer-paid model — the same structural conflict as the 2008 mortgage crisis. However, S&P and Moody's apply their own EBITDA adjustments (capitalizing operating leases, including pension deficits), often rating leverage 1–2 turns higher than company-reported figures.
FDD / QoE Team
Buy-side Financial Due Diligence
Incentive: Reality-check EBITDA
Hired by the buyer to independently verify. Theoretically the last line of defense. However, kill-deal pressure is real — damaging relationships with sellers and IBs. Typical haircut: 10–30%. KPMG data: in 35% of deals, EBITDA misses FDD estimates by 15%+ within 12 months.
The Originate-to-Distribute Problem
LevFin banks structure and price a leveraged loan, collect 1–2% in upfront fees, then sell the paper to CLOs and institutional investors — offloading credit risk within weeks. With no skin in the game post-syndication, the incentive to challenge aggressive EBITDA assumptions evaporates. A 2019 Federal Reserve working paper documented that "pro-forma" EBITDA at deal close exceeded actual LTM EBITDA by a median of 18–28% across large LBO transactions between 2014–2018.
The "One-Time" Test
The central battlefield in every EBITDA negotiation is simple: is this cost genuinely non-recurring?If it is, excluding it gives a fair picture of normalized earning power. If it is not, exclusion inflates the basis for your valuation multiple. Below is the practical spectrum FDD teams apply.
✅ Generally Accepted
Genuine one-time legal costs
Single litigation settlements, costs for specific disputes (excluding habitually litigious sectors)
Owner compensation normalization
Excess compensation above market rate at owner-operated businesses — savings realizable post-PE acquisition
IPO / deal-related advisory fees
IB and legal fees for the current transaction — structurally non-recurring
Natural disaster losses
Uninsured losses from floods, fires, and similar one-off events
⚠️ Contested — Requires Evidence
Restructuring charges
Appearing for 2+ consecutive years → FDD rejection. Valeant ran 8 straight quarters of 'non-recurring' restructuring
CEO transition / strategic review costs
Labeled one-time, but 'transformation programs' routinely run 3–4 years
COVID normalization (2020–2022)
Pandemic loss add-backs — wide variation by sector, rampant abuse in practice
New location pre-opening costs
Structurally recurring at growth-stage businesses — identical to WeWork's pattern
❌ Routinely Rejected by FDD
Stock-Based Compensation (SBC)
Buffett: 'If compensation isn't an expense, what is it? Where does it go if not to P&L?' — economic dilution is real
Unrealized synergy projections
Adding back cost savings not yet achieved. Rejected without run-rate evidence
Full SG&A exclusion (WeWork-style)
Reclassifying core operating expenses as 'non-core' — the central problem with Community-Adjusted EBITDA
Serial M&A integration costs
Integration costs from serial acquirers add-backed every deal — effectively a permanent cost
"If options aren't a form of compensation, what are they? If compensation isn't an expense, what is it? And if expenses don't go in the calculation of earnings, where in the world do they go?"
"When I see companies touting adjusted earnings that strip out most of their real costs, I take that as a signal that either management doesn't understand their own business model, or they're hoping I won't."
Case Studies
WeWork — Community-Adjusted EBITDA
Excluded all SG&A and marketing costs to flip a -$1.93B net loss into +$467M Adj. EBITDA
Think of it this way
Imagine a restaurant that says 'we're profitable if you don't count the waitstaff salaries and the advertising budget.' But those ARE the restaurant's core costs — you can't run the business without them.
GAAP Net Loss (FY2018)
-$1.93B
Community-Adjusted EBITDA
+$467M
GAAP vs. Adjusted Gap
~$2.4B
SoftBank Total Losses
-$9.5B
WeWork's business model is simple: sign long-term leases on large floors, then sublet them as short-term flexible desks. To fill those desks, you need a sales team, marketing spend, and pre-opening costs for each new location. These were WeWork's fundamental operating costs.
In its S-1 IPO filing, WeWork defined 'Community-Adjusted EBITDA' by removing not just interest, taxes, and D&A — but also all SG&A ($933M in FY2018) and pre-opening costs, labeling them 'community investments.' The result: a GAAP net loss of -$1.93B transformed into a reported profit of +$467M. A $2.4B swing.
The problem is straightforward: to sign one more member, you need sales and marketing. Strip that cost and a mature, full location looks profitable. But the business as a whole — which requires constantly filling new locations — can never be profitable on those terms.
The IPO was pulled in September 2019. WeWork filed for Chapter 11 bankruptcy in November 2023. SoftBank's total losses reached -$9.5B.
Key Takeaway
If the cost disappears, so does the business. Excluding a core operating expense can make any company look profitable on paper.
Source: WeWork S-1 filing, Aug 14, 2019 (SEC EDGAR)
Valeant — Recurring 'Non-Recurring' Charges
Restructuring costs labeled 'non-recurring' for 8 consecutive quarters — turning GAAP EPS -$0.18 into 'Cash EPS' +$10.17
Think of it this way
Imagine taking your car to the mechanic eight quarters in a row, and each time telling your spouse 'don't worry, this is the last repair bill, it won't happen again.' By the eighth time, it's not a one-time expense — you just have a broken car.
GAAP EPS (FY2015)
-$0.18
'Cash EPS' (adjusted)
+$10.17
Adjusted EBITDA (FY2015)
~$5.7B
Stock Peak → Trough
$262 → <$10
Valeant's strategy was simple: acquire pharmaceutical companies, raise drug prices, cut R&D. Because it was constantly acquiring, it had constant integration and restructuring costs — every quarter, without fail.
Each quarter, Valeant labeled these costs 'non-recurring' and excluded them from adjusted EBITDA. The effect: GAAP earnings per share of -$0.18 became the company-reported 'Cash EPS' of +$10.17. A difference of over $10 per share.
The fatal flaw: this 'one-time' charge appeared for eight consecutive quarters between 2013 and 2015. When something happens eight times in a row, it isn't a one-time event — it's a structural cost of doing business. Academic research backs this up: Doyle, Jennings & Soliman (2013) showed that 65–70% of items companies call 'non-recurring' reappear in subsequent periods.
The SEC opened an investigation in 2016. Financial statements were restated. The stock collapsed from a peak of $262 to below $10.
Key Takeaway
When the same 'one-time' item appears two or three quarters in a row, stop calling it one-time. It's a structural cost of that business.
Source: Valeant Annual Report 2015; SEC Comment Letters (EDGAR); Doyle et al. (2013), JAE
MBK Partners × Homeplus — Korea's Largest PE Buyout
Sold the stores' real estate to raise cash, then leased them back — but the new rent bill never went away
Think of it this way
Imagine selling your house to raise a lump sum of cash, then renting it back from the new owner and saying 'look how much cash we have now.' The sale proceeds are one-time. The monthly rent you now owe is permanent.
Acquisition Price
KRW 7.2T
Sale-Leaseback Proceeds
KRW 4T+
Leverage at Acquisition
~9–10× EBITDA
Court Receivership Filed
Feb 2025
In 2015, MBK Partners acquired Homeplus from Tesco for KRW 7.2T — Korea's largest-ever PE buyout, largely debt-financed at roughly 9–10× EBITDA leverage.
After the acquisition, MBK raised over KRW 4T by selling Homeplus store buildings and leasing them back (sale-leaseback). The proceeds were used to pay down debt and return capital to investors. On the surface, this looked like smart financial engineering.
But here's what changed: before the leasebacks, Homeplus owned its stores and had no rent. After the leasebacks, it had permanent, ongoing rent obligations for every store it had sold. This new cost was not fully priced into the deal's EBITDA projections.
Compounding the problem, Korean discount retail was being devastated by Coupang, Naver, and convenience stores between 2015 and 2020. At 9–10× leverage, there was no buffer to absorb this structural shift. Homeplus filed for court receivership in February 2025.
Key Takeaway
One-time sale proceeds cannot be treated as recurring EBITDA. When you sell and leaseback, a permanent rent expense is created — and it must be fully subtracted from normalized earnings.
Source: Homeplus court filing (2025); Korea Fair Trade Commission; MBK Partners press releases
HDC Hyundai Development × Asiana Airlines
Heavy maintenance costs treated as 'one-time' — COVID pulled the trigger, but the structural vulnerability was already there
Think of it this way
Imagine buying a car and the dealer says 'every five years you'll need a full engine overhaul.' If you then record that overhaul as a 'one-time expense,' you're fooling yourself. Five-year cycles are predictable — they're not surprises.
Agreed Deal Value
KRW 2.5T
Contract Signed
Dec 2019
IFRS 16 Lease Liabilities
KRW 2.4T+
Deal Outcome
Terminated (2020)
HDC Hyundai Development agreed to acquire Asiana Airlines for KRW 2.5T in December 2019. When COVID-19 grounded international flights in early 2020, HDC invoked the MAC (Material Adverse Change) clause and sought to exit. COVID was the trigger — but the structural vulnerabilities had been there from the start.
Every commercial aircraft requires a 'D-Check' (heavy maintenance) roughly every 6–8 years: a complete disassembly and inspection of every component. Cost: hundreds of millions of won per aircraft. Because these checks don't happen annually, they're often labeled 'non-recurring' and excluded from EBITDA. But for any airline that keeps flying, D-Checks are as inevitable as fuel costs — they just happen on a longer cycle.
The second issue: IFRS 16 accounting required Asiana's aircraft operating lease obligations (over KRW 2.4T) to be recognized as balance-sheet liabilities. Whether these were fully incorporated into the enterprise value calculation became a central dispute.
The deal was ultimately terminated, confirmed by court ruling. The episode became a textbook example in Korean M&A circles of the gap between adjusted EBITDA and economic reality.
Key Takeaway
'Infrequent' is not the same as 'one-time.' If a cost is predictable and inevitable — even on a 6-year cycle — it belongs in normalized EBITDA.
Source: Asiana Airlines SEC filings; HDC Hyundai Development press releases; FSS Korea
FDD: Quality of Earnings Bridge
Financial Due Diligence (FDD) teams rebuild EBITDA from scratch. The output is a "Quality of Earnings" bridge — a waterfall that walks from the seller-reported figure to the FDD-adjusted number. This single document is often the pivotal input for final price negotiation.
Reported EBITDA (Seller)
$120M
(-) Restructuring add-backs reversed
recurring pattern, 3 years
-$8M
(-) SBC add-backs reversed
economic dilution is real
-$5M
(-) Synergy projections not achieved
zero run-rate evidence
-$4M
(-) COVID normalization haircut
sector outlook risk
-$4M
FDD Adjusted EBITDA
-$21M vs. seller
$99M
-17.5%
EBITDA haircut in this example
$120M → $99M
-$315M
Deal price impact at 15× multiple
$21M × 15× = $315M
35%
Deals where FDD EBITDA misses by 15%+ within 12 months
KPMG (2022)
Red Flag Checklist
Adj. EBITDA margin significantly exceeds GAAP operating margin
Gap over 5–10% points warrants explanation
Same 'non-recurring' items appear in 2+ consecutive reporting periods
The Valeant pattern — structural cost masquerading as one-time
Add-back list exceeds 8–10 items
Complexity itself is a red flag; Doyle et al. (2013) find quality degrades sharply beyond this threshold
Synergy add-backs with no integration plan
Future savings with no evidence of execution or run-rate should be fully rejected
EBITDA headline in press release, GAAP net loss buried
Disclosure hierarchy signals management's priority — and its credibility
Adj. EBITDA used in debt covenants without independent definition
When the borrower defines the metric, covenant-lite loans lose their disciplining function
Serial acquirer's integration costs add-backed every deal
If you acquire every year and add-back integration costs every year, it is core operating cost
IFRS 16 / ASC 842 lease reclassifications without clear disclosure
Lease capitalization can materially inflate EBITDA while increasing enterprise value and debt simultaneously
Summary
Adjusted EBITDA is not fraud — until it is.
The concept has a legitimate purpose: to show normalized earnings power by stripping genuine noise. The abuse potential is structural, because every party upstream of the buyer has a financial incentive to push the number higher.
EV = Multiple × EBITDA means EBITDA adjustments are amplified at the deal level. A $10M shift becomes a $100–200M+ swing in price depending on the multiple.
The key test is not the label — it's recurrence. If the same 'one-time' item appears in three of the last five years, it is a structural cost.
Every party except the FDD team (and sometimes even the FDD team, under kill-deal pressure) has an incentive to tolerate inflated EBITDA. Structural analysis of incentives is a prerequisite for reading any adjusted disclosure.
FDD Quality of Earnings adjustments typically range from 10–30% haircuts. A 15% haircut at a 15× multiple reduces deal value by 2.25× the adjustment amount in absolute terms.
Regulatory pressure is rising. SEC Reg G and C&DI 2016 updates require equal or greater prominence for GAAP metrics vs. non-GAAP, specific labeling, and prohibition on exclusions that mislead reasonable investors.
Regulatory Framework
SEC Regulation G (2003)
Requires any public company presenting non-GAAP metrics to include a quantitative reconciliation to the most directly comparable GAAP measure. Applies to all earnings releases and investor presentations.
C&DI Updates — May 2016
Prohibited non-GAAP income metrics that exclude charges or liabilities requiring cash settlement. Explicitly prohibited presenting non-GAAP on a per-share basis without equally prominent GAAP disclosure.
Item 10(e) of Regulation S-K
Requires that non-GAAP measures not be presented more prominently than GAAP, must use accurate description labels, and must not omit material GAAP metrics. Enforced through SEC comment letters.
ASC 718 (FASB)
Stock-Based Compensation standard requires SBC to flow through the GAAP income statement. Companies that add back SBC in adjusted metrics are explicitly excluding a GAAP-mandated expense recognized because it has economic substance.
Related Concepts
References
- [1]Doyle, J., Jennings, J., & Soliman, M. (2013). Do managers define non-GAAP earnings to meet or beat analyst forecasts? Journal of Accounting and Economics, 56(1), 40–56.
- [2]WeWork Companies Inc. Form S-1, Securities and Exchange Commission (Aug 14, 2019). SEC EDGAR.
- [3]Black, E. L., Christensen, T. E., Ciesielski, J. T., & Whipple, B. C. (2018). Non-GAAP reporting: Evidence from academia and current practice. Journal of Business Finance & Accounting, 45(3–4), 259–294.
- [4]SEC Compliance & Disclosure Interpretations (C&DI): Non-GAAP Financial Measures (Updated Apr 2018). U.S. SEC.
- [5]KPMG. (2022). Global M&A Trends: Post-Deal Performance and Due Diligence Findings Survey. KPMG International.
- [6]S&P Global Ratings. (2023). Corporate Methodology: Ratios and Adjustments. S&P Global Market Intelligence.
- [7]Valeant Pharmaceuticals International, Inc. Annual Report on Form 10-K (2015). SEC EDGAR.
- [8]Lys, T., Naughton, J. P., & Wang, C. (2015). Signaling through corporate accountability reporting. Journal of Accounting and Economics, 60(1), 56–72.
- [9]Deloitte. (2023). Quality of Earnings: What You Don't Know Can Hurt You. Deloitte M&A Institute.