What the Price Implies

What the Price Implies

At $7.94 on 7 August 2026, Weibo's market capitalisation is $2.13 billion. Its own balance sheet carries $5,807.7 million of cash, investments, deposits and loans against $1,863.6 million of debt — $3,944.2 million net, or roughly twice the whole market value. Crediting that book at nothing, the price is 4.6 times the cash the advertising business generates by itself. What follows sets out what recovery on those financial assets, and what rate of decline in the operating business, the price is consistent with.

The price and the share count behind it

Weibo had 245,549,858 ordinary shares outstanding at 31 December 2025 — 157,727,834 Class A and 87,822,024 Class B [1]. At $7.94 that is $1,949.7 million of equity value. The $2.13 billion market capitalisation quoted above rests instead on the 268,560 thousand diluted shares used in reported earnings per share, a count that sits 23.0 million above the shares actually issued because the 2030 Convertible Notes are treated as converted [2]. Those notes convert at 72.6929 ADSs per $1,000 of principal, about $13.76 per ADS [3], so at $7.94 they are debt in substance and cash in prospect — the reason the maturity schedule in Claims on the Cash treats the December 2027 put as a cash obligation. Both counts appear below where they matter; the choice does not change any conclusion, because it moves the equity value and the debt figure in the same direction.

Sources: market capitalisation and trailing free-cash-flow yield from reported financials at the 7 August 2026 close; book value of $3,920.7 million and the balance-sheet inputs from the FY2025 Form 20-F, Consolidated Balance Sheets [4]; dividend of US$0.61 per share declared in March 2026 [5].

Book value per share is $15.97 on the $3,920.7 million attributable to Weibo's shareholders, so the shares change hands at roughly half of stated book [6]. Reported earnings give a trailing multiple of 4.7 times the $1.70 of FY2025 diluted earnings per share [7], and that multiple is the least reliable of the three. FY2025 net income carried a $76.7 million pick-up from equity-method investees and $21.3 million of fair-value gains; one quarter later the same lines ran the other way — a $35.0 million fair-value loss and $22.1 million of equity pick-up losses cut net income attributable to shareholders to $34.7 million from $107.0 million, and diluted earnings per share to $0.14 from $0.41, while income from operations was flat at $110.9 million [8]. A price-to-earnings ratio on this company is in large part a ratio on the investment portfolio's mark for the period.

Two balance sheets, one price

The more informative decomposition separates the assets that produce advertising revenue from the assets that produce interest, dividends and marks. Nine lines on the FY2025 balance sheet are financial rather than operating.

Sources: cash, short-term investments, amount due from SINA, long-term investments and the three debt lines from the FY2025 Form 20-F, Consolidated Balance Sheets [9]; wealth management products, related-party and third-party loans from the balance-sheet components note [10]; equity value derived from 245,549,858 shares at the 7 August 2026 close.

The debt figure is the sum of $745.6 million of unsecured senior notes, $323.9 million of convertible senior notes and $794.0 million of long-term loans, all of them carried at the Cayman parent [11]. Netting it against the financial assets at face value leaves $3,944.2 million, or $16.06 per share against a $7.94 quote. The residual line is what remains for a business that earned $464.8 million of operating income on $1,757.2 million of revenue in FY2025 [12]: negative $1,994.5 million.

Face value is the wrong mark, and the reconciliation in Claims on the Cash is why. Of that book, $946.0 million is credit extended to related parties rather than cash held, $620.5 million sits in wealth management products, and $1,663.3 million is a long-term investment portfolio whose measurement-alternative sleeve is already carried far below cost. More constraining still, the cash is in the wrong entity: the parent that owes every dollar of the debt held $286.4 million of cash and $16.5 million of short-term investments at year-end, against group cash of $2,298.9 million [13]. Moving money out of China costs at least the 5% treaty withholding rate, and the 20-F's own illustration shows RMB100 of pre-tax earnings inside the VIEs reaching a shareholder as 67.5 [14]. Financial assets that stay inside the onshore entities contribute their interest income to consolidated earnings; they do not reach an outside holder at carrying value.

Solving for the assumption

The useful move is to treat the recovery rate on that financial book — whose seven-year deployment record and marks are set out in The Investing Record — as the unknown, and read off what the price implies for the operating business at each level. The operating business's own cash generation comes from stripping the financial income out of free cash flow: FY2025 free cash flow of $477.1 million, less $116.8 million of interest income and $8.5 million of dividends received from equity-method investees, plus $75.3 million of cash interest paid on the debt that is being netted separately, gives roughly $427 million [15] [16]. That is an approximation before tax effects: the interest income removed was taxed and the interest expense added back was deductible, and the two are close enough in size that the residual error is small against the ranges below.

Source: derived from the FY2025 balance sheet and cash-flow statement [17] [18], at a market value of equity of $1,949.7 million.

Two readings follow. Writing the entire $3.94 billion down to zero — every dollar of onshore cash, every loan to SINA, the whole investment portfolio, with the debt still repaid in full — still leaves an operating business bought at 4.6 times its own annual cash generation. Crediting the book at 50 cents on the dollar leaves the operating business at approximately nothing. Across the plausible range there is no recovery assumption at which the price looks demanding on FY2025 numbers.

The operating business on its own terms

That leaves the second variable: what a flat-to-declining annuity is worth. The advertising base has not grown since 2019 and the arithmetic of its slope is set out in Advertising Base; consensus has FY2026 revenue growth at 2.6% and a forward free-cash-flow yield of 18.0%. The table below prices a perpetuity at a range of discount rates and constant decline rates, for comparison with the 4.6 times paid at zero recovery.

Source: derived, as the reciprocal of the discount rate plus the decline rate; the cash-flow base is the US$427 million computed above from the FY2025 cash-flow statement [19].

At a 12% discount rate, 4.6 times corresponds to a perpetual decline of about 10% a year; at 15%, about 7%. Weibo's advertising revenue excluding Alibaba fell 3.9% in FY2025, from $1,381.9 million to $1,327.8 million [20], and by low single digits in each of the two years before that [21]. The price at zero recovery therefore embeds a faster erosion than the company has yet delivered, and does so while assuming the entire financial book is worthless.

The read this supports: the shares are priced for a combination of outcomes — near-total impairment of the financial assets together with an acceleration of the advertising decline — that the FY2025 record does not show. The strongest fact against it is that neither leg has to be literally true for the price to be right. If the onshore cash and the related-party loans are never distributed and never realised by an outside holder, their worth to that holder approaches zero regardless of carrying value; and a business whose average daily users fell from 260 million in December 2024 to 252 million in December 2025 [22], on a margin that is drifting, can compound toward a high-single-digit revenue decline without any single dramatic year. What would change the read in either direction is observable: execution of the buyback, a refinancing of the 2027 maturities that does not consume the offshore pocket, and two consecutive quarters of positive constant-currency advertising growth without an event base.

What has actually reached shareholders

Between 2023 and 2025 Weibo paid out $200.1 million, $194.4 million and $195.6 million in dividends [23], and in 2026 a further $150 million on the $0.61 per share declared in March [24]. Four declarations of $0.85, $0.82, $0.82 and $0.61 total $3.10 per share — 39% of the current price, returned in four years [25].

Sources: cash dividends paid for 2023-2025 from the FY2025 Form 20-F, Consolidated Statements of Cash Flows [26]; the 2026 figure from the Q1 2026 earnings call [27].

The trend in that series is the more important fact. The 2026 declaration is 26% below each of the prior three, in a year when free cash flow was $477.1 million and the payout absorbed under a third of it. Set against the parent-level maturities, the cut reads as a decision about the offshore pocket rather than about earnings.

The repurchase record cuts against the value case more directly. The board authorised $200 million in December 2025, running to 31 December 2026, and the 20-F states plainly that no shares were repurchased in 2025 [28]. At $7.94, $200 million would retire roughly 25.2 million shares, 10.3% of the count. The precedent is not encouraging: the previous programme, authorised at $500 million in March 2022, spent $57.7 million buying 3,055,759 ADSs at an average of $18.88 before lapsing [29]. The company bought at $18.88 and has bought nothing at $7.94. Asked in March 2025 how it thought about shareholder returns, the CFO named two principles — driving business growth and staying competitive, and enhancing shareholder returns — and said the company was open to other measures such as buyback [30]. The authorisation followed nine months later; no execution has been disclosed in the record to August 2026. Nor has any insider bought in the open market, a point the People record sets out in full.

Nasdaq closes reported in the 20-F itself run $9.55 at end-2024 and $10.22 at end-2025 [31], against $7.94 in August 2026 and a mean sell-side target of $9.04 across sixteen analysts. The discount has widened while the operating numbers were roughly flat.

Against the listed peers

Weibo is inexpensive relative to the listed peers named in its own filings [32] [33], not only relative to its history.

Source: third-party market data as at 10 August 2026, not from company filings; iQIYI is omitted, having no meaningful trailing multiple. Earnings quality differs across the group — Weibo's denominator carries the investment-portfolio marks described above.

The comparison is indicative rather than decisive: none of these companies runs an identical model, as the peer caveats in Competition set out, and a trailing multiple on Weibo's reported earnings inherits all the volatility of its equity-method book. The narrower point survives those caveats — the discount is not a China-listing discount alone, since Autohome and Hello Group face the same jurisdiction and trade at 16.8 and 9.2 times earnings against Weibo's 5.6.

What would move the range

Four things carry most of the movement, each observable in a filing.

The 2027 maturities. Up to $1,130 million of principal can fall due in 2027 [34] against $302.9 million of liquid assets at the parent [35]. A refinancing on terms that leaves the offshore pocket intact removes the largest single reason to haircut the financial book; funding it by upstreaming, at the 5% withholding toll, quantifies the leakage instead.

Buyback execution. A disclosed repurchase converts the $200 million authorisation from an intention into per-share arithmetic, and at the current price it is the most accretive use of offshore cash available to the company.

The advertising slope. The perpetuity table is most sensitive to the decline rate: the difference between a flat base and a 5% annual decline at a 12% discount rate is 8.3 times against 5.9 times — a swing of about $1.0 billion on $427 million of cash flow, roughly half the current market value.

The tax base. Preferential treatments — high and new technology enterprise status and the research-and-development super-deduction — were worth $37.0 million in FY2025, an effect of $0.15 on basic earnings per share [36], and the filing notes the preferential rate is temporary in nature and subject to qualification [37].

Sources: Q1 2026 and Q1 2025 unaudited results [38] [39]; FY2025 free cash flow from the Consolidated Statements of Cash Flows [40].

Limits of this arithmetic

Three are worth naming. The perpetuity grid is a comparison device, not a model: it holds the decline rate and the margin constant, and Weibo's operating margin has moved from 26.9% to 28.2% and back to 26.5% across the last three years. The recovery table nets financial assets against debt as though both sat in one place, which the parent-versus-group split shows they do not — a more exact treatment would discount the onshore assets for time and tax while holding the parent's obligations at par, which pushes every recovery figure lower. And none of this rests on a control premium or a take-private: the voting structure documented in People means closing any discount is not a decision an outside shareholder can force.