China: Avoid outsourcing conviction to headlines
Megan Ie
Senior Equity Analyst
Overview
Two weeks on the ground firmed my view: China is being misread. Choppy tape, brittle confidence; yet the meetings, the queues, and the way Boards talk about cash tell a different story. No longer an index bet, it’s a stock-picker’s market where the edge belongs to businesses that prove the unit maths, return cash, and compound per share.
I see the drawdown and the bruised sentiment but, I also see a gap between price and process big enough to reward patience. Here’s why we’re staying in and what the evidence says.
What changed in 2025
Corporate sentiment snapped back but with a monetisation filter
After DeepSeek’s reasoning-model splash, nearly every deck had an “AI paragraph”- some substantive, some ceremonial. Either way, it lifted confidence and, crucially, nudged conversations toward workflow paybacks: how AI lowers service costs, trims headcount in support functions, or upsells a member base.
Policy and activity raised the floor
Early-year activity data stabilised, and policy nudged Boards to better allocate capital. That doesn’t cancel macro risk; it sets a lower bound under guidance and brings dividends/buybacks into serious conversation.
Demand is alive but is ruthless on value
The consumer isn’t absent, they’re discerning. Menus are slimmer, promos sharper, but “consumption from the heart”(IP-anchored novelty, experiences, athleisure, pets and medical aesthetics) still pulls spend. Pricing power persists where the value proposition is felt, not inferred.
Why it matters: The turn in early-2025 wasn’t a sugar high. It was a pivot from “why bother” to “show me the unit economics.” The investable set isn’t “China beta”, it’s cash compounders with governance discipline, and they exist in scale.
Chart 1: Two different varietal menus with one message - value up (1)
Three key reflections investors aren’t hearing enough
Involution vs. innovation: where returns will actually come from
On the ground, involution is visible in consumer pricing and product churn. Demand exists, but it’s exacting. Consumers will spend on IP-anchored novelty (blind-box collectibles, blocks), experiences (domestic travel), athleisure, pets, and medical aesthetics, but they’ll punish weak value propositions. The arms race of launches and promos means it’s harder for incremental marketing or capex to lift per-share outcomes. The line I heard too often - “there’s nothing I really want to buy” - isn’t about a lack of wallets, it’s fatigue with look-alike products.
Innovation differentiates business models that lift revenue density (per square metre or per member). One domestic hotel operator we met is effectively turning rooms into retail showrooms, cross-selling sleep products through a large loyalty base. That’s productivity innovation we can underwrite, it's margin-accretive and hard to copy at scale.
Shop floor automation is getting smarter. China’s humanoid /robotics push is real but already crowded. Teams are converging on similar sensor stacks, and parts lists are transparent enough that scale alone doesn’t guarantee margin. In that world, the profit pools sit in the bottlenecks: precision gear reducers, hydraulics, and high-spec bearings. We favour domestic suppliers in those choke points, where the customer payback is clear and repeatable..
Why it matters: Innovation that improves unit economics drives ROE improvement. The recovery path is through pricing power, mix, and productivity with less feature churn and more monetisation of real frictions.
China’s high quality businesses consistently deliver superior ROE vs macro sensitive peers -> innovation + cash discipline drive shareholder returns.
Chart 2(2): MSCI China Index-Weighted ROE: high quality group vs macro-related group(3)
You don’t need perfect macro to make money
Tariffs
Meetings have moved past “deal/no-deal” and towards the operative question: who eats the tax? In one Tier-1 components meeting, the US buyer absorbed most of the levy because switching costs were higher than the tariff (or cost of it) itself. That’s revealed preference about moats and substitutability. Policy is still moving, Washington’s 301 review lifted rates in targeted areas (solar inputs) and Brussels imposed five-year countervailing duties on Chinese-made EVs, but the question to ask is company specific: pass-through, mix, and customer stickiness.
Property overhang
The sector’s adjustment still weighs on beta. It doesn’t preclude stock-level winners. Companies with narrow moats, cash discipline and demand drivers unrelated to property (IP licensing, factory automation, premium experiences) continue to earn premium multiples. The trick is to track structural recovery signals and look through the daily property headlines.
Why it matters: As global flows rotate and index weights under-reflect China, dispersion is the feature, not the bug. In under-owned markets, active capital gets paid.
Chart 3: Macro dragged sectors index weight <85%/ 75% of the universe respectively: MSCI China and CSI 300 (4)
Chart 4: China, still universally unloved – GEM fund’s under allocation vs MSCI Weight (5)
From high growth to growth-and-yield
A quiet realignment around cash discipline and cash returns is underway. Policy and boardroom behaviour are meeting in the middle. The State Council’s “new Nine” (April 2024) and CSRC follow-ups tightened supervision and pushed for better dividend behaviour, while SASAC reinforced “market value management” at central SOEs. Meanwhile, Hong Kong buybacks hit fresh records through 2024 as Boards leaned into depressed multiples. None of this makes a mediocre business good, but it does shorten the distance from “cheap” to “re-rateable” for the ones that earn their keep.
Why it matters: If you’ve spent years criticising China for starving shareholders, 2025 is the year you can start asking for it in writing: ranges for payouts, logic for buybacks, and the hurdle rates that govern both.
Chart 5: All China listed universe - Use of cash (6)
Turning noise into returns: Where we’re owning it (and why)
The common thread in our holdings is simple: clear cash generation, identifiable pricing or network advantage, and management teams willing to write down (in plain English) how excess cash gets back to owners. We engage on that last mile.
Examples of where that’s led us:
1. Online recruitment
A two-sided network where matching quality beats coupon wars. High switching costs, high incremental margins, and low reinvestment needs ideal for rising owner returns over time. Management’s willingness to return surplus cash supports the thesis that we get paid while the network compounds.
2. Global PCs/devices with a rising services slice
A global brand that’s using services/software attach to lift margins and smooth the hardware cycle. The self-help levers of services penetration, recurring support, device-as-a-service are measurable. As the services mix rises, we see scope for a structural re-rating on steady operating margins.
3. Online travel
A category leader with brand and scale in a more consolidated industry. The model is asset-light, free cash flow is robust, and the business has levers in higher margin accommodation, cross-border expansion and disciplined capital returns. With capital intensity low, ongoing buybacks and a clearer payout path are natural outcomes.
Chart 6: Further value unlock via engagement (6)
Closing thought
China is too large, too interconnected and too innovative to ignore, especially at today’s starting points. You don’t need perfect macro. Own quality businesses that prove the unit maths, find Boards aligned with all shareholders interests, and the patience to sit through the volatility stretch while compounding works. That’s the faith we’re keeping.
Performance has been lacklustre. The work hasn’t been. The on-the-ground evidence from 2025 says there are more of these businesses than consensus admits. That misread between valuation and fundamentals is the opportunity.
References
(1) GIB AM EMAE team
(2) Factset, Morgan Stanley
(3) High quality group defined as strong earnings growth and ROE delivery, typical sectors like Internet, IT. Macro related group defined as sectors highly sensitive to macro cycle like property, energy, materials etc.
(4) CSIndex, Bloomberg, data as per 8 Aug 2025
(5) EPFR country allocation, Factset, Morgan Stanley, data as per June 30, 2025.
(6) Wind, FactSet, Goldman Sachs Global Investment Research
(7) GIB AM EMAE Analysis, July 2025
This content should not be construed as advice for investment in any product or security mentioned. Examples of stocks are provided for general information only to demonstrate our investment philosophy. Observations and views of GIB AM may change at any time without notice. Information and opinions presented in this document have been obtained or derived from sources believed by GIB AM to be reliable, but GIB AM makes no representation of their accuracy or completeness. GIB AM accepts no liability for loss arising from the use of this presentation. Moreover, any investment or service to which this content may relate will not be made available by GIB to retail customers.
