Chinese equities are conditionally investable, but “China” is not an investment case. Large-cap A-shares are mainly a governance-and-profit-recovery trade; growth A-shares depend more heavily on technology becoming profitable; offshore platforms rely on internet and technology cash flows. The case works only when profit grows, poor capital allocation does not consume the value per share, and price, currency and product costs still leave enough return.
01 · The thirty-second answer
If you read only one section, read this one. The rest of the note explains these four answers.
Yes, but it must be separated
Chinese equities are not one asset. Large caps, growth companies and offshore platforms have very different holdings, valuations and risks. One macro view cannot settle all three.
Per-share earnings, not GDP
Economic growth reaches shareholders only through corporate profit, capital allocation, dividends or effective buybacks. Policy support and industrial success are not enough.
Profit follows reform
Better rules can reduce value destruction, but they cannot make companies profitable. Our base case needs better governance and sustained growth in earnings and cash return per share.
Cheap is not safe; expensive can still rise
Our three-year base scenarios are about +6% a year for large caps, +6% for A-share growth and +9% for offshore technology. Bear paths are roughly −9%, −19% and −14%. These are stress tests, not forecasts.
02 · What can we actually buy?
An index is not a miniature Chinese economy. It is a basket selected by a set of rules. Choose the businesses first, then the vehicle.
CSI 300, CSI A500 and SSE 50 cover large financial, industrial, consumer, healthcare and technology companies. They are the cleaner expression of governance, dividends and broad profit recovery.
ChiNext and STAR 50 are more concentrated in advanced manufacturing, healthcare and hard technology. More upside depends on commercial success; valuation compression can also produce larger losses.
Tencent, Alibaba, Meituan and JD.com are primarily listed in Hong Kong or the US. They add internet profits that A-shares omit, together with listing-structure, policy and currency risk.
An index is not the final return. ETF domicile, fees, tracking error, currency conversion, hedging and tax can all change what an Australian investor receives.
Different names do not guarantee diversification: at 30 June 2026, the constituent-weight overlap between CSI A500 and CSI 300 was about 76.96%. We calculate it by summing the smaller weight for each shared stock. It is a dated snapshot, not a permanent property.
03 · How does a shareholder make money?
For an Australian investor, long-run return is roughly: dividends + per-share earnings growth + valuation change + holding-currency movement against AUD − product costs − tax. Trading currency and economic currency exposure are not always the same.
- Does the company earn durable profit and cash?
- After dividends, buybacks, reinvestment and new shares, how much value remains per share?
- Does the chosen index and product own the profit pool we actually want?
- After valuation, currency, fees and tax, is the return worth the risk?
| Check | For non-financial companies | For financial companies | Common mistake |
|---|---|---|---|
| Profit | Margins, free cash flow and return on invested capital | Net interest margin, credit cost, capital and book-value return | Treating industry scale as shareholder value |
| Per-share value | EPS, net issuance, cash-covered dividends and disciplined buybacks | Book value per share, capital needs and distributable profit | Assuming every dividend or buyback creates extra value |
| Vehicle | Holdings, weights, rebalancing, fees and tracking | The same checks, including bank and insurance weights | Reading the index name instead of its holdings |
| Price | Valuation, expected earnings, FX and opportunity cost | Valuation alongside asset quality and the credit cycle | Ignoring price because the story is right |
Profit retained inside a company is not automatically “lost”. It can create value when management reinvests at attractive returns. Dividends do not create extra total return by themselves either; their value is in limiting poor reinvestment and making cash return more dependable.
04 · What is our investment case?
Our case is not “policy has turned, therefore shares must rise”. It contains distinct investments, and each one needs profit to complete the logic.
Less value destruction, followed by profit
Changes to issuance, selling, dividends, delisting and investor protection can reduce damage to value per share. If margins, EPS and cash payouts then recover, low-to-moderate valuations can provide a conditional return.
Technology must become a high-return business
Local substitution, research spending and production are industrial evidence, not return evidence. High valuations need revenue, margin, cash flow and returns on capital to improve together.
Mature cash flows must offset structural risk
Platform business models and cash flows are often more mature, but policy, listing structure, concentration and valuation cycles differ. Filling a gap in A-shares does not remove price discipline.
GDP growth, policy support, long-term capital inflows and technology substitution are background conditions. We do not count them as return until they become higher index EPS, cash return or a more attractive entry price.
05 · What has happened, and what remains unproven?
New rules are a starting point, not an investment result. We separate observed facts from the claims that still need evidence.
| Question | What we have seen | What remains to be proved |
|---|---|---|
| Market governance | The 2024 State Council programme and related rules tightened listing, dividends, selling, delisting and investor protection. | Whether enforcement persists and actually reduces poor financing and capital allocation. |
| Cash return | Policy gives more weight to stable dividends and disciplined buybacks. | Whether dividends are cash-covered and buybacks are sensibly priced, retired and not reissued through compensation. |
| Index renewal | Indices rebalance; A-shares are not permanently fixed in an old-economy mix. | Whether new constituents produce more durable per-share earnings instead of merely adding fashionable sectors. |
| Profit recovery | Markets now carry sharply different valuations and earnings expectations. | Whether property, weak demand, price competition and excess capacity allow sustained EPS and capital-return improvement. |
| Long-term capital | Policy encourages insurers, pensions and other long-duration investors. | Whether inflows improve governance and price discovery, or simply raise valuations before earnings improve. |
Stronger delisting can improve long-run market quality while causing immediate losses for existing owners. Tighter listing rules improve future securities, not the present index overnight. Reform must pass through enforcement, rebalancing and company economics.
06 · What return might today's price support?
We start with official valuation snapshots, then use three representative indices in a three-year stress test. The purpose is not to forecast an index level. It is to make the win and loss assumptions visible.
| Market proxy | Trailing P/E | Dividend yield | What the price needs |
|---|---|---|---|
| SSE 50 | 11.78× | 2.71% | Stable large-cap profit, dividends and governance |
| CSI 300 | 14.36× | 2.23% | Broad large-cap earnings recovery |
| CSI A500 | 16.23× | 1.88% | Broader industry representation and faster growth |
| STAR 50 | 77.94× | 0.25% | Very high technology commercialisation and profit growth |
| ChiNext | 38.79× | Not shown in snapshot | Growth sectors reaccelerate while holding margins |
| Hang Seng TECH | 29.76× | 1.00% | Double-digit platform and technology earnings growth |
ChiNext uses its August 2026 official factsheet; the other snapshots are at 31 July 2026. Holdings, earnings and valuation all change, so these figures describe a historical starting point only.
| Representative index | Bear | Base | Bull | Approx. annualised return |
|---|---|---|---|---|
| CSI 300 14.36× / 2.23% yield | EPS −3% a year; exit at 12×; CNY/AUD −3% a year | EPS +5%; exit at 14×; FX flat | EPS +10%; exit at 17×; CNY/AUD +2% | −9% / +6% / +21% |
| STAR 50 77.94× / 0.25% yield | EPS flat; exit at 45×; CNY/AUD −3% | EPS +15%; exit at 60×; FX flat | EPS +25%; exit at 80×; CNY/AUD +2% | −19% / +6% / +29% |
| Hang Seng TECH 29.76× / 1.00% yield | EPS flat; exit at 20×; HKD/AUD −3% | EPS +10%; exit at 28×; FX flat | EPS +18%; exit at 35×; HKD/AUD +2% | −14% / +9% / +28% |
How to read the table: it starts from 31 July 2026 valuations, holds for three years and adds the starting dividend yield as a simple approximation with reinvestment. It excludes specific ETF fees, tracking error and tax. The scenarios have no reliable probabilities and cannot be averaged into one expected return. The real tail can be worse.
Large caps are not expensive, but the base path is about 6% a year—below our roughly 9% long-run hurdle. Growth indices need very high earnings growth to offset lower valuation and have the largest bear loss. Offshore technology comes close to the hurdle in the base case, with higher structural risk and upside. There is no single “China” answer that can replace vehicle choice and price discipline.
07 · What makes the investment win or lose?
An economy or industry can succeed while its shares lose. Earnings, value per share and entry price determine the outcome together.
Reform reduces damage and profit follows
Index EPS and cash return improve for several quarters; net issuance stays controlled; technology spending earns high returns; valuation does not prepay the result; and currency and product costs do not consume the gain.
The direction is right, but price moves first
Governance and company performance improve, yet long-term capital raises valuations before earnings arrive. The market rises and companies get healthier, but shareholders earn only an ordinary market return.
Policy supports price while profit fails
Revenue and output rise alongside price wars, low returns and repeated financing; offshore structures face new restrictions; and earnings, valuation and AUD return fall together. China can still grow while the shares lose.
Large-cap A-shares and offshore platforms solve different coverage problems and can be treated as separate decisions. Growth A-shares are closer to a high-volatility option on commercial success. We first look through China exposure already held in global funds, then decide whether any deliberate gap remains.
08 · How risks hurt return, and what we watch
| Priority | Risk | How shareholders are hurt | What we watch |
|---|---|---|---|
| 1 | Profit never follows reform | Property, weak demand, price wars and excess capacity suppress earnings, leaving valuation recovery unsupported. | Four consecutive quarters of index EPS, non-financial cash flow, financial asset quality and margins. |
| 2 | Financing and allocation dilute value | Total profit grows, but new shares, low-return investment or uncovered dividends prevent equal growth per share. | Net issuance, retired buybacks, dividend cover, returns on capital and enforcement of selling rules. |
| 3 | Technology succeeds but owners do not | Substitution and output rise through subsidy, falling prices and heavy investment, producing poor financial returns. | Margins, operating cash flow, subsidy dependence, revenue from R&D and capital returns. |
| 4 | Offshore structure and geopolitics | Regulation, audit access, VIEs, export controls, sanctions or listing changes hit earnings expectations and valuation together. | Listing rights, cash transfer, enforcement, fund terms and cross-border market access. |
| 5 | Price, currency and vehicle | Profit growth is consumed by lower valuation, weaker CNY or HKD, fees, tax and tracking error. | Valuation versus earnings, CNY/AUD and HKD/AUD, all-in product cost and portfolio overlap. |
If index EPS and genuine cash return fail to improve for four consecutive quarters while net issuance rises, we would remove the governance benefit from our base return. If price rises far faster than earnings, we would treat return as prepaid. A material change in product rights, cross-border access or listing structure would require a fresh investability decision, not merely a lower valuation multiple.
09 · Sources, method and notice
This note was updated on 2 September 2026. Valuation snapshots are mostly at 31 July 2026; the CSI A500–CSI 300 overlap is at 30 June 2026. Official factsheets update with prices and rebalances, so dates are retained to avoid presenting changing data as permanent facts.
- State Council capital-market programme and CSRC dividend and delisting explanation.
- CSI 300 factsheet, CSI A500 factsheet and SSE 50 factsheet.
- STAR 50 factsheet and ChiNext factsheet.
- Hang Seng TECH factsheet and CSI A500 methodology.
Our estimates: the three-year scenarios, overlap calculation and interpretations of index drivers are not forecasts from the index providers. The scenarios combine EPS growth, exit P/E, dividends and currency against AUD, and exclude product-specific tax and costs.
This note is general research and information only. It is not investment or financial product advice, a recommendation, an offer or solicitation. Past performance is not a guide to future results.