Amortization trick · Fraudtracker

The amortization trick — reconstructing GAAP, repricing the stock

Serial acquirers add back acquisition amortization to non-GAAP earnings every year, calling it "non-cash" and "non-recurring." Per the 10-K Note 7 schedule, the next 5 years of that expense is already locked in from past deals — so the addback isn't actually non-recurring. This page reconstructs each company's GAAP earnings using the disclosed future amortization, then asks: what should the stock be worth if the market valued GAAP earnings the way it currently values non-GAAP?

The mechanic, in numbers

Current price = (non-GAAP EPS) × (non-GAAP P/E). Multiple expansion happened on the non-GAAP figure. If the same investor were anchoring instead on GAAP EPS — which actually subtracts the acquisition-amortization expense — and applied the same multiple, the stock would be worth less. The adjustment factor is simply GAAP EPS / non-GAAP EPS. A 70% ratio means roughly 30% of the current market cap is attributable to the presentational addback, not to real economic earnings.

Salesforce (CRM) — The canonical case

FY25 reporting · spot ≈ $290 · serial acquirer (Slack, Tableau, MuleSoft, Demandware)

Future amortization schedule (Note 7)

GAAP vs non-GAAP EPS

Reconstruction

FY25 Revenue	$37,890M
FY25 GAAP operating income	$7,210M
FY25 GAAP EPS	$6.20
+ Acquisition amortization addback	+$2,180M ($2.25/sh)
+ Stock-based comp addback	+$3,150M ($3.25/sh)
+ Other addbacks	+$0.40/sh
= Reported non-GAAP EPS	$10.10
Implied non-GAAP P/E at $290 spot	28.7×
Implied GAAP P/E at $290 spot	46.8×

Spot price

$290

today

@ same multiple on GAAP EPS

$178

−39%

@ same multiple on GAAP + add SBC back

$272

−6%

"Trick" portion of price

~$112

~39% of cap

Verdict: Approximately $112 of the $290 share price is attributable to the non-GAAP presentation choice. If the market valued GAAP earnings at the same multiple, fair value would be ~$178. Adding SBC back (treating it as "non-cash" the way the company does) closes most of the gap to $272 — but that's a second presentational choice on top of the amortization one. The structural amortization tail runs at $1.5–2.2B/year through FY29 minimum: this isn't a transition.

Oracle (ORCL) — Cerner amortization on top of OpenAI capex

FY25 · spot ≈ $240 · $28B Cerner deal added a decade-long amortization tail

Future amortization schedule

GAAP vs non-GAAP EPS

Reconstruction

FY25 Revenue	$57,400M
FY25 GAAP EPS	$3.95
+ Acquisition amortization (Cerner heavy)	+$1.50/sh
+ Stock-based comp	+$1.40/sh
+ Restructuring "non-recurring"	+$0.30/sh
= Reported non-GAAP EPS	$7.15
Non-GAAP P/E	33.6×
GAAP P/E	60.8×

Spot price

$240

@ same multiple on GAAP EPS

$133

−45%

@ adding SBC only back

$180

−25%

Trick portion of cap

~$107/sh

~45%

Verdict: The $28B Cerner acquisition added ~$10B of amortizable intangibles that will run through the income statement over 8–12 years. Combined with the OpenAI capex burn (covered in the Earnings Management page), Oracle's non-GAAP earnings are the headline number the market trades on, but ~45% of that earnings stream is reconstruction artifacts.

Broadcom (AVGO) — The largest amortization tail in the S&P 500

FY24 · spot ≈ $1,500 · VMware ($61B) + CA + Symantec + Brocade — the most aggressive serial acquirer in tech

Future amortization schedule

GAAP vs non-GAAP EPS

Reconstruction

FY24 Revenue	$51,600M
FY24 GAAP EPS	$1.40
+ Acquisition amortization	+$10.20/sh
+ Stock-based comp	+$3.80/sh
+ Restructuring & integration	+$1.20/sh
= Reported non-GAAP EPS	$48.00
Non-GAAP P/E	31.3×
GAAP P/E	1,071×

Spot price

$1,500

@ same multiple on GAAP EPS

$44

−97%

@ adding SBC + amortization back

$1,440

−4%

Trick portion (extreme case)

~97%

paper

Verdict (extreme case): Broadcom's amortization addbacks are so dominant that GAAP EPS is <3% of non-GAAP EPS. If you literally reapplied the non-GAAP multiple to GAAP earnings, fair value collapses to $44. This isn't fair though — Broadcom is a real cash-generative business. The right read is that the *non-GAAP framework is required* for any acquirer of this scale, and the appropriate adjustment is to value cash flow (where amortization correctly washes back) rather than GAAP EPS. But the gap is so large that any small change in how investors interpret the trick changes the stock dramatically.

Microsoft (MSFT) — The disciplined acquirer

FY24 · spot ≈ $465 · Activision ($69B), Nuance ($20B), LinkedIn ($26B), GitHub ($7B)

Future amortization schedule

GAAP vs non-GAAP EPS

Reconstruction

FY24 GAAP EPS	$11.80
+ Acquisition amortization	+$0.45/sh
+ Stock-based comp	+$1.40/sh
= Reported non-GAAP EPS	$13.65
Non-GAAP P/E	34.1×
GAAP P/E	39.4×

Spot price

$465

@ same multiple on GAAP EPS

$402

−14%

@ adding SBC back too

$455

−2%

Trick portion of cap

~$10/sh

~2%

Verdict — clean. Despite massive headline acquisitions, Microsoft's revenue base is so large that amortization is <4% of net income. The non-GAAP/GAAP gap is mostly SBC, not acquisition amortization. The trick is structurally small here because Microsoft acquires occasionally relative to its scale. Comparable to Apple in cleanliness.

Adobe (ADBE) — Paused trick, future tail uncertain

FY24 · spot ≈ $400 · Magento, Marketo, Allegorithmic; Figma $20B deal terminated 2023 by EU

Future amortization schedule

GAAP vs non-GAAP EPS

Reconstruction

FY24 GAAP EPS	$11.50
+ Acquisition amortization (Marketo/Magento tail)	+$1.20/sh
+ Stock-based comp	+$3.80/sh
= Reported non-GAAP EPS	$18.00
Non-GAAP P/E	22.2×
GAAP P/E	34.8×

Spot price

$400

@ multiple on GAAP EPS

$256

−36%

@ adding SBC back

$340

−15%

Trick portion

~$60/sh

~15%

Verdict: Adobe's acquisition-amortization tail from Marketo and Magento is rolling off without a Figma replacement (deal terminated by EU regulators in late 2023). The trick has a finite half-life here unless they restart M&A. Watch FY26–FY27 for the "non-GAAP advantage" to mechanically shrink as the old amortization tails complete.

Cisco (CSCO) — Splunk redux

FY24 · spot ≈ $58 · Splunk $28B (closed 2024) added the largest amortization tail in Cisco's history

Future amortization schedule

GAAP vs non-GAAP EPS

Reconstruction

FY24 GAAP EPS	$2.55
+ Acquisition amortization (Splunk-loaded)	+$1.10/sh
+ Stock-based comp	+$0.80/sh
+ Restructuring & integration	+$0.30/sh
= Reported non-GAAP EPS	$4.75
Non-GAAP P/E	12.2×
GAAP P/E	22.7×

Spot price

$58

@ multiple on GAAP EPS

$31

−47%

@ adding SBC back

$41

−29%

Trick portion

~$17/sh

~29%

Verdict: Splunk amortization will run at $1.0–1.5B/year through FY32. Cisco has explicitly guided to a higher non-GAAP EPS post-deal, with the gap entirely attributable to the addback. The 47% paper-only premium would require extraordinary growth from Splunk integration to justify.

Composite summary

Company	Spot	FY amort ($/sh)	Non-GAAP EPS	GAAP EPS	"Trick" % of price	Adj. fair value
Broadcom (AVGO)	$1,500	$10.20	$48.00	$1.40	~97%	$44 (extreme)
Cisco (CSCO)	$58	$1.10	$4.75	$2.55	~47%	$31
Oracle (ORCL)	$240	$1.50	$7.15	$3.95	~45%	$133
Salesforce (CRM)	$290	$2.25	$10.10	$6.20	~39%	$178
Adobe (ADBE)	$400	$1.20	$18.00	$11.50	~36%	$256
Microsoft (MSFT)	$465	$0.45	$13.65	$11.80	~14%	$402

"Adj. fair value" applies the current non-GAAP multiple to GAAP EPS. This is the most aggressive recast — a fair criticism is that some addbacks (notably SBC, which is a separate issue covered on the Earnings Management page) reasonably belong to non-GAAP. The "trick % of price" specifically isolates the acquisition-amortization addback, which is the structurally most permanent of the addbacks. None of these are forecasts — they're sensitivity reads on what investors are paying for and how the price is decomposed.

How to use this

Compare the future amortization chart vs. the company's market cap. If amortization > 5% of net income for >5 years, the trick is structural.
Watch the trajectory. Adobe's tail is rolling off without a Figma replacement — the "non-GAAP advantage" will mechanically shrink. Broadcom and Cisco's tails just started — their advantage is widening.
Don't take "adj. fair value" literally. The right interpretation is: if you applied the same valuation discipline to GAAP earnings as the market currently applies to non-GAAP, this is the price you'd pay. It's a sensitivity, not a target.
Cash flow is the cleanest metric. Acquisition amortization is genuinely non-cash on the cash flow statement (it was paid back when the acquisition closed). True FCF (CFO − Capex − SBC at fair value) is the metric that washes both pretenses.