JPMorgan: Alibaba's Share Placement Selloff Overdone; Three Tech Giants Show Divergent AI Financing Paths
Nashnova编辑部
Alibaba raised US$10.3 billion through a share placement, yet lost roughly US$19 billion in market cap the next day — JPMorgan argues the market overshot, repricing the scale and duration of AI capex rather than the placement itself.
The stock dropped 8.5% — how much can the placement actually explain?
Alibaba placed 710 million shares at an 8.4% discount, raising HK$80 billion.
The stock fell 8.5% the next day; Tencent shed 2.2% and the Hang Seng Tech Index lost 2.1% in sympathy.
This means → JPMorgan calculates the placement discount accounts for only about 0.3% of the decline. After stripping out sector contagion, the abnormal repricing totalled roughly US$19 billion — twice the capital raised.
In plain terms = for every 100 dollars raised, about 185 dollars of market cap vanished. Even if the money were set on fire, the loss should cap at ~US$10 billion. The market was pricing something else.
How does JPMorgan break down the economics of this deal?
The bank applies four tests. ① Compute capex earns an IRR of 22%, pays back in roughly two years and eleven months, and shows an NPV of RMB 36 per share at a 10% discount rate — broadly matching management's "sub-3-year payback, >13% blended return" claim, assuming 60% goes to infrastructure.
② Does the placement boost EPS? At 80%+ infra allocation, yes — from year two onward. At the 60% base case, FY2028 EPS dilutes by about 0.5%, then turns roughly neutral.
③ Is equity the cheapest funding source? No. Debt would save 3–3.5 percentage points of EPS drag — RMB bond coupons run 2%–3%, well below the 12% earnings yield.
④ Does the deal add intrinsic value per share? No. Starting from a pre-deal fair value of HK$205, per-share value declines across all scenarios.
How long does the raised capital last?
At a RMB 200 billion capex base case, Alibaba's annual internal funding gap is roughly RMB 100 billion. If June-quarter capex is annualised to RMB 270 billion, the gap widens to RMB 170 billion.
The placement covers approximately eight and a half months (base case) or five months (stress case).
This means → JPMorgan expects more debt issuance over the next 12 months. A second parent-level equity raise, however, would be read as a negative signal of deteriorating returns.
What paths are Tencent and Baidu taking?
Tencent is largely self-funding: annualised H1 operating cash flow of ~RMB 310 billion roughly matches its RMB 200 billion full-year capex forecast. It also holds a RMB 875 billion listed and unlisted investment portfolio as a buffer.
This means → Tencent's cost is implicit — foregone portfolio returns, tax friction on disposals, and market impact. No financing is needed, but the price is not zero.
Baidu has the weakest internal funding capacity: June-quarter operating cash flow was just RMB 3.4 billion against RMB 11.4 billion in capex. The gap is plugged by loans; equity fundraising sits at the subsidiary level — Kunlun Chip — diluting the parent's stake in AI assets.
How heavy is the depreciation burden, and where is the return ceiling?
One year of base-case capex generates total depreciation equal to 28% of Alibaba's FY2027 expected net income, 13% for Tencent, and 23% for Baidu.
Yet incremental EBITDA — earnings before interest, taxes, depreciation and amortisation, i.e. operating profit before non-cash charges — exceeds depreciation from the first full year. The real risk lies in utilisation rates and pricing, not the depreciation load itself.
This means → returns have a ceiling: 80% utilisation + 10% price cut → ~11% return; 70% utilisation + 20% price cut → only ~3%.
What would make JPMorgan change its mind?
Five risk triggers: ① three-year capex materially exceeds RMB 380 billion without a matching cloud-revenue upgrade; ② raised capital remains largely idle on the balance sheet beyond 12 months; ③ AI lab losses fail to decline quarter-on-quarter in September; ④ compute pricing keeps falling as supply loosens; ⑤ a second parent-level equity raise within 12 months.
JPMorgan maintains Overweight ratings on all three — Alibaba, Tencent, and Baidu. Alibaba trades at just 12x / 8x FY2027/FY2028 earnings.
In plain terms = JPMorgan's view is that the post-placement drop overshot — but if any one of those five triggers fires, that "overshot" call has to be revisited. September-quarter results are the key checkpoint.
市场有风险，内容仅供研究参考，不构成投资建议。