News Corporation (NASDAQ: NWSA) — Two World-Class Franchises Handcuffed to a Non-Voting Share and a Family Trust

Independent Equity Research — Initiation Report date: 2026-07-18 · Reference price: $28.43 (NWSA close 2026-07-17)

⚡ Claude’s Take

This block is the author’s own independent opinion and general information only — not investment advice, and not a recommendation to buy or sell any security. The analysis that follows (Sections 1–15) is presented position-free; the only view expressed is in this block.

Verdict: HOLD / accumulate-on-weakness (into the low-to-mid $20s). Conviction: medium. Framing: a contrarian sum-of-the-parts value play, structurally trapped by control — “great parts, wrong wrapper.”

News Corp is two genuinely great businesses — Dow Jones’s professional-information engine (Risk & Compliance + OPIS energy-price benchmarks + Factiva) and REA Group, one of the best digital-classifieds monopolies on the planet — bolted to a mediocre #2 US property portal, a decent-not-great book publisher, and a melting newspaper stub, and then wrapped in a non-voting Class A share whose holders (the economic majority) cannot vote at all. My honest read of the numbers: marked to market, the pieces are worth roughly $38–40 a share versus $28.43 today — a ~25–30% holdco/conglomerate discount — and, most strikingly, at the current price the market is paying only ~7–9x EV/EBITDA for all of Dow Jones, below the cheapest listed information-services comp and less than half what Thomson Reuters, S&P Global, or Moody’s command. That is cheap for a franchise management is steering toward a $1 billion EBITDA target, funded by high-margin recurring data.

So why only a HOLD? Because the one force that could close that discount — a family split that forced a recap or a Digital-Real-Estate spin — died in September 2025, when the Murdoch succession war settled and handed Lachlan Murdoch entrenched control of both News Corp and Fox. Starboard already tried: its dual-class-elimination proposal, backed by all four proxy advisors, was crushed at ~35% in November 2024 because only the family-controlled Class B votes, and it was not even re-filed in 2025. This is the textbook value-trap-by-control: you are paid to wait — via an accelerating buyback ($459M year-to-date, funded by Foxtel cash) and genuine compounding at Dow Jones and REA — but you should not underwrite the discount closing on any timetable you control. The stock sits at the bear end of its own asset-value range, so downside is asset-protected; the upside needs a catalyst that no longer exists. Buy the parts at a discount and clip the buyback, but size it as a patient value position, not a special-situation home run. The single evidence that would flip me bullish: a credible, board-endorsed structural separation of Digital Real Estate (or a dual-class collapse). The single evidence that would flip me bearish: Dow Jones organic growth or R&C/Energy retention rolling over, which would remove the one leg holding the SOTP up. Tag: the best assets in publishing, priced like the worst governance in it.

📈 Stock Price Action — Five-Year Event Map

Over five years NWSA round-tripped from a reopening high near $26 (May 2021) down to a rate-shock low of ~$14.5 (Sep 2022), then more than doubled off that low to an all-time high of $30.90 (Sep 2025) on the Foxtel exit, activist attention, and the Dow Jones/REA earnings inflection. It trades at $28.43 today, in a 52-week range of $22.31–$30.90, roughly 8% below its all-time high. The arc is a value name that de-rated on rates, then re-rated on self-help — not a momentum blow-off (the stock’s factor profile is value-tilted with negative momentum loading).

Each price move is a Fact; the attributed cause is Interpretation, cross-referenced to earnings prints, 8-K events, and the news record. No price target or recommendation is implied here — the opportunity judgment sits in Claude’s Take above.

1. Executive Summary

News Corporation is not one business; it is a holding company for four very different ones, and the gap between its best and worst assets is the entire investment question. Two segments — Dow Jones and Digital Real Estate Services — generate roughly 73% of the company’s positive segment EBITDA on just 49% of revenue, at 25% and 33% margins respectively. The other two — Book Publishing (HarperCollins) and News Media — are a quarter of revenue each but throw off comparatively little profit, and over the last nine reported months their EBITDA actually declined. The celebrated “12 consecutive quarters of segment-EBITDA growth” is a company-level statistic carried entirely by the two crown jewels.

Inside Dow Jones sits the real prize: a fast-growing Professional Information Business — Dow Jones Risk & Compliance and the OPIS/Chemical Market Analytics energy-and-chemicals price-benchmark franchise — that is a genuine wide-moat, recurring-revenue, ~90%-retention information toll of the same archetype as Thomson Reuters and S&P Global Commodity Insights. Alongside it, REA Group (61.4%-owned, ASX-listed) is a network-effects near-monopoly in Australian property classifieds — four times the traffic of its nearest rival — with pricing power that grows yield in the mid-teens independent of listing volumes.

The problem is threefold. First, consolidated returns are poor: blended ROIC is ~5.5%, below any reasonable cost of capital, because two high-return franchises are diluted by a subscale #2 US portal (Move/Realtor.com, no durable moat — a housing-cycle option) and a structurally declining newspaper business. This is a sum-of-the-parts story, not a compounder. Second, the parts are worth far more than the whole: marked to market, our base SOTP is ~$38–40 per share against $28.43, a ~25–30% discount; the market implies a value for all of Dow Jones of roughly 7–9x EV/EBITDA — cheaper than the cheapest listed information-services peer. Third — and decisively — that discount is structurally entrenched by control. Class A shares (NWSA) do not vote; the Murdoch family, through LGC Holdco, controls the only voting class (Class B/NWS), capped at 44% of the vote. Starboard’s 2024 campaign to collapse the dual-class structure, endorsed by all four major proxy advisors, failed at ~35% and was not re-filed in 2025. The September 2025 succession settlement cemented Lachlan Murdoch’s control of both News Corp and Fox and extinguished the last plausible catalyst for a structural simplification.

Capital allocation has improved markedly in the last three years — the Dow Jones data acquisitions (OPIS, Chemical Market Analytics, Oxford Analytica, Dragonfly, Eco-Movement) were well-chosen and well-priced, the Foxtel exit was correct, and the buyback has accelerated sharply — but that improvement is young and was funded largely by Foxtel proceeds rather than a durable change in the underlying return profile. The balance sheet is a fortress (net cash excluding leases; total debt/EBITDA ~2.1x; interest coverage ~14x), which underwrites the buyback and the optionality. On balance, News Corp is a collection of excellent and poor assets, correctly cheap on the whole because control keeps the discount from closing — a value opportunity for a patient owner, and a value trap for an impatient one.

2. Business Overview

News Corporation was created in June 2013 when the old News Corporation split into two: 21st Century Fox (the film/TV/pay-TV assets, later mostly sold to Disney; the residual is today’s Fox Corporation) and the “new” News Corporation — the publishing, information, and digital-real-estate assets that are the subject of this memo. The company reports on a June 30 fiscal year. Following the divestiture of the Foxtel pay-TV business to DAZN (closed April 2025, now presented as discontinued operations), News Corp operates four continuing segments plus a corporate “Other” line.

Fiscal 2025 (ended 6/30/25): revenue $8,452M; total segment EBITDA $1,415M (16.7% margin). The segment composition is the single most important fact about the company:

Dow Jones produces The Wall Street Journal, Barron’s, MarketWatch, and Investor’s Business Daily on the consumer side, and — increasingly the profit engine — a Professional Information Business (PIB): Dow Jones Risk & Compliance, Dow Jones Energy (OPIS, Chemical Market Analytics), Factiva, and Dow Jones Newswires. Roughly 84% of Dow Jones revenue is now digital.

Digital Real Estate Services is two assets: REA Group (61.4%-owned, separately listed on the ASX), whose flagship realestate.com.au dominates Australian property classifieds and which also owns REA India and Mortgage Choice; and Move/Realtor.com in the United States, the #2/#3 property portal behind Zillow.

Book Publishing is HarperCollins, the world’s #2 consumer book publisher, where backlist titles are ~64% of consumer revenue.

News Media is the legacy newspaper and digital-masthead business — The Sun and The Times/Sunday Times (News UK), the New York Post (and the newly launched California Post), The Australian and the News Corp Australia mastheads, and the talkSPORT radio franchise.

Revenue quality. Circulation and subscription revenue was ~$3.0B of the $8.45B total (~36%), and real-estate subscription revenue adds more; on our read roughly half the company is recurring, and the recurring half is disproportionately the profitable half. The cyclical/transactional remainder — advertising (chiefly News Media and Dow Jones), book unit sales, and Realtor.com lead-generation — is where the volatility and the weakness sit.

Verdict: A four-part holding company whose economic center of gravity — Dow Jones and REA — is genuinely high-quality and recurring, wrapped around two lower-quality businesses that flatter the top line and drag the return profile. The structure, not any single segment, is the story.

3. Industry Dynamics

News Corp straddles four industries with sharply different structural attractiveness.

(1) B2B financial information & compliance data (Dow Jones PIB) — structurally the best industry the company touches. This is a classic oligopoly of recurring-revenue “toll roads”: Bloomberg, LSEG/Refinitiv, S&P Global, Moody’s, MSCI, FactSet, Verisk, and — in the energy/chemical price-assessment niche — S&P Global Commodity Insights (Platts), Argus Media, and ICIS (RELX). Barriers to entry are high: proprietary datasets, deep archives, workflow embedding, and — for benchmark businesses — the self-reinforcing property that once a price assessment is written into counterparties’ contracts, it cannot easily be displaced. Secular tailwinds are real: the compliance addressable market (Risk & Compliance) is roughly $3.7 billion growing 11–13%, driven by ever-expanding sanctions, anti-money-laundering, and KYC obligations; the energy-data market is being reshaped by the surge in US LNG and crude exports, where OPIS has unique domestic coverage. Pricing power and ~90% retention are the industry’s fingerprints. Verdict: structurally attractive, secularly growing — the crown-jewel industry.

(2) Online real-estate classifieds — bifurcated: superb where #1, brutal where subscale. The economics are winner-take-most: the portal with the most buyers attracts the most listings, which attracts more buyers, and the leader monetizes via “depth” (agents paying to make listings more prominent) and yield increases that outrun transaction volumes. In Australia, REA is the #1 and enjoys those monopoly-grade economics. In the United States, the same industry structure works against News Corp, because Move/Realtor.com is the subscale challenger to Zillow — it has audience but not lock-in. Regulation is a live variable (the NAR commission settlement is reshaping US agent economics), and the housing cycle governs transaction-linked revenue. Verdict: excellent (REA) and poor (Move) within one segment — rank is destiny.

(3) Trade book publishing — mature, concentrated, margin-capped. The “Big Five” (Penguin Random House, HarperCollins, Hachette, Macmillan, Simon & Schuster) hold most of the market; the failed Penguin Random House–Simon & Schuster merger (blocked 2022) left the field fragmented enough that HarperCollins keeps a stable #2 slot. But Amazon’s buyer power caps margins at ~13–14%, and the model is hit-driven and advance-intensive. Verdict: structurally stable but not attractive.

(4) Newspapers (News Media) — structurally declining. This is the same secular collapse documented across the sector: print advertising captured by Google and Meta, local news decimated, and a digital-subscription transition that only a handful of titles (chiefly The New York Times) have decisively won. News Corp’s mastheads are managing decline — extracting cost efficiencies (the News UK/DMG print partnership; Australian streamlining) and converting readers to digital subscriptions (The Times 676k; News Corp Australia 1.2M digital) — but they are not escaping it. Verdict: structurally declining; a managed run-off.

Applying the Marathon capital-cycle lens: the professional-information and Australian-classifieds industries are ones where high returns have not attracted enough new capacity to compete them away (network effects and benchmark lock-in defend the incumbents), while newspapers are the opposite — a shrinking profit pool where capital has rationally exited. That divergence is exactly why the SOTP framing matters more here than the blended P&L.

4. Competitive Position

Dow Jones Professional Information Business — a wide, durable moat (switching costs + regulatory-/benchmark-embedded intangibles). This is the one franchise in the portfolio that is unambiguously great, and it is the one the market appears to value least. Risk & Compliance grew ~15% in FY25 to ~$337M and +19% in the March 2026 quarter; Dow Jones Energy grew ~11% to ~$278M. The moat mechanism is concrete and financially visible: R&C data is embedded in customers’ regulated anti-bribery, sanctions, and KYC workflows, so switching means re-validating a compliance process — expensive and risky, which is why retention runs ~90%. OPIS and Chemical Market Analytics are reference-price benchmarks for energy and base chemicals — the direct analog to Platts (S&P), Argus, and ICIS (RELX) — and a benchmark that is written into supply contracts is among the deepest moats in all of information services. Named competitors from the 10-K include Reuters, RELX/LexisNexis, Refinitiv/World-Check, S&P Global, Argus, and DTN. This is a Thomson-Reuters-adjacent franchise of the same class (recurring, high-gross-margin, high-retention), if smaller and less mature. Verdict: durable wide moat. One material caveat: OPIS’s benchmark integrity is under legal scrutiny — PVC-pipe antitrust class actions, a DOJ Antitrust grand-jury subpoena (January 2025), and a state civil investigative demand over the PVC/Pipe Weekly Report. Because a benchmark’s value is its perceived integrity, this is a genuine, if contained, moat risk to monitor.

WSJ / Barron’s (Dow Jones consumer) — moderate, brand-based moat, AI-pressured. The Wall Street Journal has real brand equity and demonstrated pricing power (digital full price raised to $44.99 for new subscribers), but it is a discretionary consumer product whose markets/how-to/evergreen content is precisely what LLM answer-engines and Google’s AI overviews commoditize. Verdict: moderate moat, structurally pressured.

REA Group — a wide, network-effects near-monopoly. REA is one of the best digital-classifieds businesses in the world, in the tier of Rightmove (UK). realestate.com.au draws ~132M average monthly visits and ~4x the traffic of its nearest competitor. The two-sided network effect — agents must advertise where every buyer looks; buyers go where every listing is — converts into pricing power via depth upgrades and mid-teens yield growth independent of listing volumes (REA revenue +20% reported / +8% constant-currency in the March 2026 quarter on a +14% yield). The distant #2, Domain, needed CoStar’s deep pockets to stay relevant — itself a testament to REA’s dominance. Verdict: durable wide moat.

Move/Realtor.com — a subscale #2, not a durable moat. Realtor.com has real audience (~72M monthly uniques; 31% visit share, ~6x Homes.com and ~3x Redfin per Comscore in Q3 FY26) but sits behind Zillow’s network effects; its listings are ~94% commodity MLS-sourced and its very brand is licensed from NAR rather than owned. It monetizes lead-generation and listing subscriptions in a market where the #1 captures the network economics. This is the mirror image of REA within the same segment — audience without lock-in. Verdict: no durable moat; a housing-cycle option rather than a franchise (with genuine upside optionality: on a trailing-12-month basis, Realtor’s revenue-to-existing-home-sales ratio is >20% higher than at the 2022 housing peak, so a transaction recovery would drop through powerfully).

HarperCollins — narrow, shallow moat (scale + backlist intangibles) in a modest industry. #2 globally; backlist (~64% of consumer revenue) provides some annuity quality, and author/agent relationships plus distribution scale are real advantages over small presses. But margins are ~13–14%, Amazon caps pricing, and the model is hit-driven — the March 2026 quarter was strong (EBITDA +14% on the Heated Rivalry TV tie-in) but nine-month FY26 EBITDA still fell. Verdict: a good #2 in a not-great industry.

News Media — brands, but no durable moat. The Sun, The Times, the New York Post, and the Australian mastheads carry brand recognition, but there is no mechanism that would let profits persist against the secular decline; the digital-subscription transition is a mitigation, not a moat. Verdict: no durable moat.

5. Growth History and Forward Opportunities

News Corp’s growth splits cleanly along the same fault line as its quality. The two crown jewels are growing organically and profitably; the other two are cyclical or declining.

Dow Jones has posted 13 consecutive quarters of year-over-year EBITDA growth, driven by the Professional Information Business (PIB revenue +7% in FY25 to ~$903M, now ~39% of Dow Jones circulation-and-subscription revenue), pricing on the consumer side, and digital-advertising recovery. Management’s Investor Day (March 2026) set a target of $1 billion in annual Dow Jones segment EBITDA within five years (from ~$588M in FY25), to be reached via R&C (riding an ~11–13%-growth TAM), energy-data expansion (new benchmarks, US-export tailwind), Factiva, and consumer-yield initiatives (WSJ at $44.99; enterprise partnerships lifting digital net adds). This is high-quality, largely organic growth.

Digital Real Estate growth is REA-led: REA revenue +9.5% and EBITDA +15.6% over the nine months to March 2026, on yield/depth rather than volumes (Australian new-buy listings were roughly flat). Realtor.com is the option: core real-estate revenue +15% in the March 2026 quarter despite existing-home sales near historic lows (3.98M annualized in March 2026), with adjacencies (new homes, rentals, sellers) now ~22% of revenue — positioned to inflect when mortgage rates and transaction liquidity normalize.

Book Publishing growth is hit-driven and was negative on a nine-month basis; News Media is in managed decline. Neither is a forward growth driver.

The AI-licensing “opportunity” deserves a skeptical parse (Section 4’s caveat expanded): the durable, moat-consistent piece is licensing Dow Jones data (OPIS benchmarks, R&C, Factiva) — an input AI models genuinely need and cannot hallucinate. The more fragile piece is news-content licensing and litigation settlements: the OpenAI and Meta partnerships have uncertain renewal economics, and the headline $1.5B Anthropic settlement (of which HarperCollins/News Corp receives a share starting ~late CY2026) is a one-time event, not run-rate revenue. Management’s “News Corp is an AI-inputs company” framing is roughly 70% real when applied to Dow Jones data and roughly 70% marketing when applied to the mastheads. Treat AI licensing as a real but partly non-recurring bridge, not a reason to re-rate the newspaper stub.

Verdict: high-quality organic growth at ~half the company (Dow Jones, REA), cyclical optionality at Realtor, and secular decline at News Media. The aggregate grows — total-company EBITDA +10% over nine months — but it does not compound in the return sense, because the incremental capital and the blended base earn below the cost of capital.

6. Financial Quality

Quality of earnings — start by throwing out the headline EPS. FY2025 GAAP diluted EPS of $2.08 is badly distorted by a $692M discontinued-operations gain on the Foxtel disposal; continuing-operations diluted EPS was only $1.14. Likewise, the FY22→FY23 revenue drop from $10,385M to $8,012M is an optical artifact of reclassifying Foxtel to discontinued operations, not an organic collapse. Any valuation lens anchored on GAAP EPS or on the full-history revenue line is mismeasuring the company — which is precisely why the own-history P/E percentile (near the bottom of its range, “cheap”) is misleading here and should be ignored in favor of P/S and the sum-of-the-parts.

Margins and their trajectory. Total segment EBITDA margin has climbed from 13.6% (FY21) to 16.7% (FY25), and to 15.7% in the March 2026 quarter (up 130bps YoY) — genuine operating leverage, but concentrated: it comes from Dow Jones (25% margin) and Digital Real Estate (33%), while News Media runs at 7% and HarperCollins at ~14%. The mix shift toward the high-margin franchises is the real margin story.

Cash generation. Free cash flow was ~$978M in FY25 (~$1.72/share), ~$897M in FY24, and ~$777M in FY23 — capex-light, subscription-and-data-driven. FCF conversion is solid, and the business is not capital-hungry (capex modest, though management flagged “moderately higher” capex in FY26).

Returns on capital — the crux of the “not a compounder” verdict. Consolidated ROIC was ~5.5% in FY25 (and 4.1% FY24, 3.1% FY23), ROE and ROA similarly depressed once the Foxtel gain is stripped. These are below any reasonable WACC — the blended entity does not earn its cost of capital, because REA’s and Dow Jones’s high returns are diluted by goodwill-heavy acquisitions, the subscale Move, and the declining News Media. The moats are real at the segment level; they simply do not show up in the consolidated return statistics, which is the mathematical signature of a conglomerate that should be valued on its parts.

Balance sheet — a fortress. As of 6/30/25: cash $2,403M against gross debt of $2,940M (of which ~$1.14B is finance leases). On a headline basis net debt is ~$537M; treating finance leases as operating (defensible, since segment EBITDA already reflects the leased assets) puts the company at ~$0.6B net cash. Total debt/EBITDA is ~2.1x, net-debt/EBITDA is ~−0.4x, and interest coverage is ~14x. A new $1.5B unsecured revolver (March 2026) provides ample liquidity. This fortress balance sheet is what funds the buyback and preserves the optionality embedded in the discount. (Note: some data aggregators render a negative book value per share for NWSA; the correct book value is ~$15.3/share on $8.77B of equity ex-minority, and ~$5.4/share on a tangible basis.)

Verdict: economics improve with scale at the segment level (Dow Jones and REA get better as they grow), but the consolidated entity earns below its cost of capital — the definition of a business worth more broken up than blended.

7. Capital Allocation

Recently good, but young and Foxtel-funded. The three-year record is genuinely improved, and the improvement is concentrated in two places.

The good — the Dow Jones data pivot. Management bought OPIS from S&P/IHS for $1.15B (Feb 2022), Base Chemicals/Chemical Market Analytics for $295M (2022), and the tuck-ins Oxford Analytica and Dragonfly (Risk & Compliance, 2023) and Eco-Movement (EV-charging data, 2023). These built the recurring, high-margin, data-moat Professional Information Business at defensible prices — the engine behind the $1B DJ-EBITDA target. This is intelligent, on-strategy M&A.

The good — the Foxtel exit. Selling Foxtel to DAZN (closed April 2025, A$3.4B EV) removed a low-margin, capital-hungry, structurally declining pay-TV asset. News Corp received ~$592M in shareholder-loan cash repayment plus a ~6% (non-cash, illiquid, minority) equity stake in DAZN. Correct exit of the worst business in the portfolio.

The mixed legacy. The crown-jewel 61.4% REA stake and the 2014 Move/Realtor.com acquisition ($950M) define the Digital Real Estate segment — one brilliant, one subscale. Against that sit FY2020 impairments of ~$1.5B+ and prior write-downs (Amplify education, Australian print) — a reminder that this management team’s earlier capital allocation destroyed real value.

Shareholder returns. The buyback accelerated sharply in FY26 — $459M year-to-date (Q3 $193M) versus a modest $150M/$117M/$240M in the prior three years — against a new $1B authorization with ~$851M remaining as of March 2026. Management explicitly frames the buyback as arbitraging the NAV discount (“materially undervalued relative to net asset value”). The dividend is a token $0.20/year (~0.7% yield, ~23% of FCF), clearly subordinate to repurchases.

The honest assessment. The recent record is strong, but three caveats temper the verdict: (1) it is only ~three years old; (2) the acceleration was funded by Foxtel proceeds, not by a structural change in the return profile; and (3) it sits atop a conglomerate that still doesn’t earn its cost of capital. Repurchasing stock at a ~25–30% discount to a marked-to-market NAV is genuinely accretive and the right use of the Foxtel cash — but it is a symptom of the control-driven discount, not a cure for it. Verdict: improved and currently intelligent, but young, Foxtel-funded, and not yet proven as a durable culture.

8. Changes and Headwinds — Last Two Years

The last two years have been unusually eventful, and the events cut in opposite directions.

Structural simplification (positive): the Foxtel/DAZN sale (agreed December 2024, closed April 2025) and the accompanying reclassification cleaned up the P&L and freed cash for buybacks. The Dow Jones Investor Day (March 2026) formalized the $1B-EBITDA target and the “digital intelligence platform” repositioning.

AI monetization (mixed): a wave of licensing deals — the OpenAI partnership, a Meta agreement, Bloomberg licensing Dow Jones AI rights, and the $1.5B Anthropic settlement — alongside litigation against Perplexity. Genuinely value-additive for the Dow Jones data assets; partly one-time for the news content (Section 5).

Governance events (net negative for the equity thesis): the Murdoch succession war ran through the Nevada probate court, where a commissioner found in December 2024 that Rupert and Lachlan had acted in “bad faith” in trying to amend the irrevocable family trust — then it was settled in September 2025 (~$3.3B; the three non-aligned siblings bought out) in a way that cemented Lachlan’s control of both News Corp and Fox. Starboard’s dual-class-elimination campaign was defeated (~35%) at the November 2024 annual meeting despite unanimous proxy-advisor support and was not re-filed in 2025. A New Stockholders Agreement (September 2025) capped the Murdoch group at 44% of Class B voting power but also revealed that LGC Holdco pledged ~30.4M of its 62.6M Class B shares (~49%) as loan collateral to finance the buyout — a new forced-sale tail-risk on the control block that ISS flagged and the supervote overrode.

Cyclical/operational headwinds: US existing-home sales near historic lows (pressuring Realtor’s ceiling), high mortgage rates, News Media EBITDA pressured by the California Post launch investment and softer UK/Australian print, and the OPIS antitrust scrutiny (Section 4).

Verdict: the operational changes strengthen the thesis (cleaner portfolio, growing crown jewels, arbitraging buyback); the governance changes weaken it (control hardened, activist catalyst dead). On balance, the two years made News Corp a better business and a more entrenched discount simultaneously.

9. Risk Analysis

The dominant risk is not operational but structural: the very control that makes News Corp a “value trap” also caps the realistic downside (an asset-value floor set by the marketable REA stake and the fortress balance sheet). The risks that would genuinely break the thesis are a Dow Jones organic slowdown or an OPIS-benchmark impairment — the legs holding the SOTP up.

10. Valuation Discussion (Embedded Expectations)

News Corp must be valued on its parts; the blended multiple is meaningless because REA is consolidated at 100% of revenue but the market values it as a premium, separately-listed growth stock. The own-history valuation percentiles capture the tension exactly: P/E near the bottom of its multi-year range (optically cheap, but distorted by the Foxtel gain — ignore it) versus P/S and P/B near the top (genuinely rich on sales/book). The P/S being at a decade high is not a contradiction; it reflects that News Corp has re-rated from a cheap melting-print conglomerate (~1.0–1.5x sales historically) to ~1.9x sales today as the high-margin, high-multiple mix (REA + Dow Jones) became a larger share of the whole. Read P/S and the SOTP, not P/E.

Sum-of-the-parts (base case), marked to market:

Base SOTP ≈ $38–40 vs. $28.43 → a ~25–30% holdco discount; ~33–40% upside if it closes. A methodological note guards against the double-count trap: because the REA stake is valued at 61.4% of REA’s market cap, the $615M book minority interest (essentially the 38.6% of REA News Corp does not own) is already excluded and must not be subtracted again. (Marked to market, that REA minority is worth ~$5.7B — a measure of how much REA value the balance sheet hides.)

The embedded-expectations “killer insight” — what the market pays for Dow Jones. Back the marketable REA stake out of News Corp’s own price:

• NWSA equity market cap ≈ 565M × $28.43 ≈ $16.1B.
• Less REA stake at market (61.4%) ≈ $9.04B.
• Implied equity value of “everything else” ≈ $7.0B; add net debt → implied EV of the non-REA stub ≈ $7.6B.

Credit the other private pieces at base values (HarperCollins $2.6B + Move $1.45B + News Media $0.28B − corporate $1.12B ≈ $3.2B), and the residual implied for Dow Jones is ~$4.4B — roughly 7.2x EV/EBITDA on ~$600M. Even crediting the private pieces conservatively puts Dow Jones at only ~9x. The market is valuing all of Dow Jones — WSJ/Barron’s plus the fast-growing Risk & Compliance, OPIS energy-data, and Factiva professional-information engines — at ~7–9x EV/EBITDA, below the cheapest listed information-services comp (Gartner ~8x, decelerating) and less than half the 14–22x that Thomson Reuters, S&P, Moody’s, and MSCI command, and roughly a third of NYT’s 22.7x for the consumer-news franchise WSJ directly rivals. The stub is cheap; the market ascribes essentially no premium to the data compounder inside Dow Jones.

Scenarios:

The current $28.43 sits at the bear end of the SOTP range. Downside is asset-protected (bear ≈ current price, floored by the marketable REA stake and net cash); the asymmetry is to the upside if the discount ever narrows. The Foxtel disposal is the value-crystallizing template — but, per Section 8, the catalyst that would repeat it (a family-driven separation) was extinguished in September 2025. No price target and no recommendation — this is embedded-expectations and scenario analysis only.

11. Variant Perception

Consensus view: News Corp is a transforming, digital-first media-and-information company with two crown-jewel franchises (Dow Jones and REA), an improving margin structure, an accelerating buyback, and emerging AI-licensing optionality — cheap on a sum-of-the-parts basis but perennially discounted for the Murdoch control and conglomerate form.

The strongest bull case: The parts are worth ~$38–40 vs. $28.43, the market prices Dow Jones at ~7–9x while it marches toward a $1B EBITDA target, REA compounds at mid-teens yields, Realtor.com is a free call option on a US housing recovery, the balance sheet is net cash, and the buyback shrinks the share count at a fat discount to NAV. You are paid to wait for the discount to close — and a structural separation of Digital Real Estate, if it ever comes, is a 30%+ re-rating.

The strongest bear case: It is a value trap. Consolidated ROIC is below WACC and always will be while the conglomerate stays intact. The one catalyst that could force the discount closed — a Murdoch family split — died in September 2025; Starboard already lost the vote it could never win because Class A doesn’t vote. Two of four segments are shrinking, the AI-licensing windfall is partly one-time, OPIS faces antitrust scrutiny, and Realtor is levered to a housing market that may stay frozen. You could hold this for years, watch the assets compound, and still see the discount persist because control-block owners have no incentive to close it.

The 3–5 assumptions that matter most:

1. Does the conglomerate discount ever close? (The single biggest swing factor; governed by control, not fundamentals.)
2. Does Dow Jones hit ~$1B EBITDA, and does R&C/Energy retention hold ~90%? (The leg holding up the SOTP.)
3. Does REA keep compounding yield through the Australian housing cycle? (56% of the marked-to-market value.)
4. Is AI-content licensing recurring or one-time? (Determines whether the “AI inputs” narrative is a re-rate or a headline.)
5. Does Realtor’s housing optionality convert, or stay frozen? (The main cyclical upside lever.)

What would falsify each side: The bull is falsified if Dow Jones organic growth decelerates or OPIS integrity is impaired (removing the SOTP’s growth leg), or if the buyback slows once Foxtel cash is spent. The bear is falsified if the board endorses a Digital-Real-Estate separation or a dual-class collapse, or if a future family liquidity need reopens the structural question.

Factor-positioning cross-check (Section overlay): News Corp screens as a value-tilted, low-beta (0.76), idiosyncratic name with a negative momentum loading (−0.34) — i.e., it is not a crowded momentum trade; it is an abandoned-then-recovering value name whose recent ~+11% quarter is a self-help bounce off a weak trailing year (rs_12m −4.1). That is consistent with the contrarian SOTP framing: the tape reflects a stock the market has largely given up re-rating, which is where consensus is most likely offsides if a catalyst ever appears — and most likely correct if one never does.

12. Fact vs. Interpretation

13. Open Questions

1. Will the board ever endorse a structural separation of Digital Real Estate (REA + Move) or a dual-class collapse — and if not absent a family catalyst, is the ~25–30% discount effectively permanent?
2. Exact combined Class A + Class B share count for the per-share bridge — we use ~565M (Class A 376.8M + Class B 187.9M); confirm at the next 10-K. (The SOTP conclusion holds either way.)
3. How much of the AI-licensing revenue is recurring versus one-time settlement, and what are the OpenAI/Meta renewal economics?
4. Does OPIS’s benchmark franchise survive the antitrust scrutiny intact, and what is the settlement/reputational exposure?
5. What happens to the 30.4M pledged Class B shares if the collateral is called — does it create forced selling of the control block, and would that reopen the structural question?
6. Does Realtor.com’s housing optionality convert if/when mortgage rates fall, and how much drops to EBITDA versus reinvestment?
7. AUD/USD sensitivity: at ~0.698 the REA stake is ~$9.0B; a move to 0.65 would cut it ~$0.6B — how much of the recent NAV improvement is FX versus REA re-rating?

14. What Must Be True

For the bull case (the discount closes and the parts get their value):

• Dow Jones reaches ~$1B EBITDA with R&C/Energy retention holding ~90%, and REA keeps compounding yield — so the SOTP’s growth legs stay intact.
• A structural catalyst emerges: a board-endorsed Digital-Real-Estate separation, a dual-class collapse, or a family liquidity event that reopens the question.
• The buyback continues shrinking the count at a discount even after Foxtel cash is spent.
• Falsification test: if two more years pass with no structural action, continued ~35%-type governance outcomes, and a Dow Jones organic deceleration, the “cheap parts” thesis is dead money — the discount is permanent and the compounding leg is gone.

For the bear case (a permanent value trap):

• Murdoch control stays entrenched (it did, and hardened, in September 2025), so the discount never closes.
• Consolidated ROIC stays below WACC as the conglomerate remains intact.
• AI licensing proves largely one-time and News Media keeps shrinking, capping the blended growth.
• Falsification test: if the board announces a credible separation of Digital Real Estate or a dual-class recapitalization — or if the pledged control block forces a change — the “trapped forever” thesis breaks and the stock re-rates toward NAV.

Sections 1–15 carry no investment recommendation and no price target; the only view expressed is in Claude’s Take at the top, which is the author’s own opinion. The Source Appendix follows below.

APPENDIX A — Standard Diligence Questionnaire

News Corporation (NASDAQ: NWSA) — as of 2026-07-18

Supplemental to the research memo. Fact / Interpretation / Assumption labeled where it matters.

General

What thoughtful questions have other investors asked about this company? The dominant investor questions are structural, not operational: (1) Will the ~25–30% conglomerate/holdco discount ever close, and what would force it? (Starboard’s answer — collapse the dual-class and separate Digital Real Estate — was voted down.) (2) What is Dow Jones really worth on its own, given the market implies ~7–9x EV/EBITDA? (3) Is the “AI inputs” licensing revenue recurring or a one-time settlement bridge? (4) How much of the SOTP value is REA (answer: ~56% of market cap), and is the rest essentially free? (5) Does the buyback continue once Foxtel cash is exhausted? (Interpretation.)

Cyclicality & Earnings Nature

Are earnings at a cyclical high or low? Mixed. Dow Jones and REA are near cyclical highs (record margins), while Realtor.com/Move is near a cyclical low (US existing-home sales ~3.98M annualized, historic lows) — so blended earnings embed a depressed housing contribution with recovery optionality. (Fact + Interpretation.)

Driven by external environment or internal actions? Both: internal (Dow Jones data-M&A pivot, cost discipline, Foxtel exit, WSJ pricing) and external (rates/housing, FX, AI-licensing demand). The margin expansion is largely internal/mix-driven; the Realtor drag is external.

How stable are revenues? ~Half recurring (subscriptions, R&C/energy data, REA agent subscriptions), and the recurring half is the more profitable half. The cyclical remainder — advertising, book unit sales, Realtor lead-gen — is where volatility concentrates.

Outlook for products/services; how big is the market? Dow Jones R&C TAM ~$3.7B growing 11–13%; energy-data expanding with US LNG/crude exports; REA in a structurally growing (if cyclical) Australian classifieds market; Realtor a call option on a ~$1T+ US housing-transaction pool; News Media a shrinking market. International exposure is large (Australia via REA, UK via News UK).

Business Quality & Competitive Moat

Is the industry getting more or less competitive? Segment-dependent: professional-information and Australian classifieds are stable oligopolies/monopolies; newspapers face intensifying disintermediation (Google/Meta/AI); US real-estate portals are a fierce #1-vs-#2 contest (Zillow vs. Move).

How profitable is the business (ROIC/ROE)? Consolidated ROIC ~5.5% — below cost of capital (Fact). But segment-level economics are excellent (Dow Jones 25% margins, Digital RE 33%). This gap is the entire SOTP thesis.

How profitable is the industry; barriers to entry? High barriers in professional information (data/archive/benchmark lock-in, switching costs) and Australian classifieds (network effects); low barriers/weak economics in newspapers; capped margins in publishing (Amazon).

Can the business be easily understood? Reasonably — four discrete segments — but the consolidated financials are distorted by the Foxtel disposal and the 100%-consolidation of a 61.4%-owned REA, so it takes work to see the parts.

Undermined by foreign low-cost labor? No — content/data/brand businesses, not labor-cost-exposed manufacturing.

Do brands matter? Yes — WSJ, Barron’s, The Times, HarperCollins, realestate.com.au are genuine brand assets; brand is part of the moat for Dow Jones consumer and REA.

Nature of competition; switching costs? Highest switching costs at Dow Jones R&C (compliance-workflow-embedded, ~90% retention) and OPIS (contract-referenced benchmarks); network-effect lock-in at REA; weak/none at Move and News Media.

Financial Condition & Balance Sheet

Assets not fully recognized on the balance sheet? Yes, materially: the 61.4% REA stake is consolidated at book, not at its ~$9.0B market value; the ~6% DAZN equity stake is a non-cash minority holding. Book value understates economic value (the REA minority alone is worth ~$5.7B at market vs. $615M booked). (Fact + Interpretation.)

Off-balance-sheet liabilities? ~$1.14B of finance leases are on-balance-sheet; pension liabilities modest (~$117M). Nothing alarming. Watch the 30.4M pledged Class B shares as a governance (not corporate) liability.

How conservative is the accounting? Reasonable; management guides to non-GAAP “total segment EBITDA” and “adjusted EPS” — track the GAAP-to-adjusted bridge, especially the one-time Foxtel gain and impairment history (~$1.5B+ in FY2020).

How CapEx-hungry? Low — a content/data/subscription model; FCF conversion is solid (~$978M FCF FY25). Management flagged “moderately higher” capex in FY26.

Capital Allocation & Management

How much FCF, and how is it used? ~$978M FY25 FCF; increasingly directed to buybacks ($459M FY26 YTD, $851M authorization remaining), a token dividend ($0.20/yr), and on-strategy tuck-in M&A. Philosophy: arbitrage the NAV discount via repurchases.

Significant acquisitions recently? OPIS ($1.15B, 2022), Chemical Market Analytics ($295M), Oxford Analytica, Dragonfly, Eco-Movement (2023) — all building the Dow Jones data engine. Divestiture: Foxtel to DAZN (2025). (Fact.)

Buying back shares / issuing to insiders? Buying back (accelerating); no unusual insider issuance. SBC modest.

Compensation policy / management motivations? CEO Robert Thomson ~$18.5M target (40% adj. EPS / 40% adj. FCF / 20% relative TSR — reasonable metrics); Lachlan Murdoch is non-executive Chair (~$1.0M, including a ~$713k “other” line). The concern is not the metrics but that the comp committee answers to a controlling family (ISS’s objections were outvoted). (Fact + Interpretation.)

Valuation & Market Data

ADR, MLP, or K-1 issuer? No — NWSA is a US-domiciled common share (Class A, non-voting); no K-1.

Dividend policy? $0.20/year semi-annual (~0.7% yield), ~23% FCF payout — subordinate to buybacks.

How profitable is the business? Segment-level highly profitable at the crown jewels; consolidated returns below cost of capital (the SOTP tension).

Net income diverging from cash flow? GAAP net income is distorted upward by the one-time Foxtel gain; on a continuing/adjusted basis, FCF (~$978M) modestly exceeds continuing net income — clean cash conversion.

Risks & Downside

What would cause the stock to decline? A Dow Jones organic slowdown or OPIS-integrity impairment; an Australian housing/REA shock; a governance/pledge event; a broad multiple de-rating; or simply the market re-confirming the discount is permanent.

Risk of catastrophic loss? Very low — fortress balance sheet (net cash ex-leases, ~14x coverage) and an asset floor from the marketable REA stake.

Chance of a total loss? Negligible — asset-backed, net-cash, cash-generative.

Recent News & Events

Has the business environment changed recently? Yes: Foxtel divested (2025), Dow Jones Investor Day set the $1B EBITDA target (March 2026), a wave of AI-licensing deals and the $1.5B Anthropic settlement, and — critically — the Murdoch succession settled (September 2025), cementing Lachlan’s control and killing the activist catalyst.

Significant acquisitions / accounting changes / new markets? Dow Jones tuck-ins ongoing; California Post launched (News Media investment); no adverse accounting changes; the Foxtel reclassification is the main presentation change.

APPENDIX B — Source Appendix

News Corporation (NASDAQ: NWSA) — Sources & Evidence Register (accessed 2026-07-18)

Primary filings (SEC — CIK 0001564708, sec.gov/edgar)

• FY2025 Form 10-K (filed 2025-08-06, FYE 6/30/25) — segment revenue/EBITDA, competition, risk factors, Foxtel discontinued-ops, share classes.
• Form 10-Q Q3 FY2026 (filed 2026-05-08, qtr ended 3/31/26) — nine-month segment trends, buyback, professional-information-business growth; also Q1 (2025-11-07) and Q2 (2026-02-06) 10-Qs.
• 2025 DEF 14A proxy (2025-10-08) — dual-class mechanics, share counts (Class A 376,776,326 / Class B 187,862,061), Murdoch/LGC ownership, exec comp, related-party disclosures; DEFA14A / ISS solicitation materials (2025-10-08, 2025-11-04).
• 8-K filings (FY24–FY26) — Foxtel/DAZN sale agreement (12/22/24) and close (4/2/25); CFO transition (11/7/24); Murdoch settlement & New Stockholders Agreement (9/8–9/10/25); $1.5B credit facility (3/27/26); Investor Day (3/2026).
• Form 4 filings — reviewed for insider signal; routine grants/settlements (codes M/A/D), no code-P open-market purchases; family net seller/pledger (14M Class B secondary Sep 2025; 30.4M pledged).

Earnings call

• News Corp FY2026 Q3 earnings call (2026-05-07) — Robert Thomson & Lavanya Chandrashekar; segment detail, $1B Dow Jones EBITDA target, AI-licensing framing, buyback pace, WSJ pricing ($44.99), Realtor Comscore data (newscorp.com investor relations).

Quantitative data

• Multi-year financial statements, profitability/credit/per-share ratios, enterprise value, and valuation multiples (FY2020–FY2025), reconciled to the 10-K.
• Own-history valuation percentiles (7/17/26): P/E 15.2x (Foxtel-distorted), P/B 1.86x, P/S 1.79x.
• Five-year price history (adjusted/unadjusted OHLCV, EMAs, beta/alpha) used for the price-action event map.
• Factor/risk model: loadings (Value +0.30, Momentum −0.34, Market 0.76–0.82), risk-adjusted track record (y1 −3.3%, y3 +13.4% ann.), beta 0.76, idiosyncratic vol ~20.5%.

External / market data

• REA Group (ASX: REA) market cap A$21.10B (A$160.86, ~131M shares, 7/17/26) — stockanalysis.com/quote/asx/REA/market-cap/
• AUD/USD 0.698 (7/17/26) — tradingeconomics.com; exchangerates.org.uk
• News Corp 61.4% REA ownership — en.wikipedia.org/wiki/REA_Group; Yahoo Finance
• Starboard dual-class campaign & vote outcome — Businesswire (9/9/24, 11/6/24); Bloomberg/Hollywood Reporter (11/20/24); CNBC (10/17/23)
• Murdoch succession / Nevada ruling & settlement — CNN (9/8/25); Nevada probate ruling (12/2024); Wikipedia “Succession of Rupert Murdoch”
• OPIS antitrust — public reporting on PVC-pipe class actions, DOJ grand-jury subpoena (1/2025), state CID (4/2025)
• AI licensing — OpenAI/Meta partnership announcements; Bloomberg licensing of Dow Jones AI rights; $1.5B Anthropic settlement reporting; Perplexity litigation

Peer comparables (public multiples, as referenced)

• Thomson Reuters (TRI ~13.9x EV/EBITDA), S&P Global (~18x), Moody’s (~22x), MSCI (~27x), FactSet (~11x), RELX (~15.8x), Gartner (~8x) — Dow Jones professional-information comps.
• The New York Times (NYT ~22.7x EV/EBITDA, ~17% ROIC) — consumer-news subscription comp for WSJ/Barron’s.
• Zillow (~2.9x sales) — Move/Realtor.com comp.
• Fox Corporation (FOXA) — Murdoch-complex governance / dual-class reference.

Primary filings take precedence over aggregated/third-party data; where an aggregator and the 10-K diverge, the filing governs.