T. Rowe Price China Evolution Equity Fund: Extending China’s technology exposure beyond the mega-cap

Why add a third China technology allocation?

The GF CSI All-Share Information Technology ETF (SZSE: 159939) and the iShares Hang Seng TECH ETF (HKEX: 3067) already provide a strong indexed core across China's AI value chain. The former captures major A-share semiconductor and hardware companies, while the latter provides exposure to leading platforms, cloud computing, large language models, and application monetisation. These, however, capture large companies.

This is where the T. Rowe Price Funds SICAV – China Evolution Equity Fund (“China Evolution”) comes in — a strategy led since 2019 by Portfolio Manager Wenli Zheng, a 17-year T. Rowe Price veteran, backed by nine China-focused analysts, 13 regional specialists, and the firm's 170-strong global research team. It adds a different investment approach to this foundation. Rather than allowing market capitalisation to determine portfolio weights, the manager actively selects and sizes companies according to their expected earnings potential.

This means that the strategy searches beyond the large companies that dominate mainstream indices, allowing the fund to identify opportunities that may receive little attention or only marginal weights in passive portfolios. According to T. Rowe Price research, 99% of China’s long-term outperformers started the period with a market capitalisation below USD 30 billion. Long-term outperformers are defined as Chinese stocks that compounded at an annual rate of at least 20% between 2016 and 2026. This supports the case for looking beyond the benchmark-heavy names most indices concentrate in. It also broadens coverage beyond chips and internet platforms into materials, electronic components, and selected energy and power-infrastructure businesses supporting the AI investment cycle.

China Evolution should therefore be positioned as an actively managed satellite alongside 159939 and 3067, preserving the ETFs as the portfolio's core exposures while adding differentiated companies, industries, and active stock selection. The fund holds 64 companies, with projected earnings growth more than double that of the comparison index.

Where China Evolution sits in the AI value chain

The fund's holdings sit mainly in the first three layers of the five-layer AI value chain — the stage where capital expenditure, and the revenue it generates, tends to arrive earliest. That timing is the key distinction: some companies recognise this revenue as the spending happens, while others remain further downstream, still bearing the upfront investment before any payoff arrives.

Table 1: China Evolution, 159939, and 3067 cover different, overlapping layers of the AI value chain.

Every additional tranche of AI computing capacity requires platform companies to build power and data-centre infrastructure and to buy computing and networking hardware. That spending flows straight through as orders to upstream suppliers, lifting their capacity utilisation and demand for higher-value products.

The analogy is the gold rush: whichever miner eventually strikes gold, the companies selling the picks and shovels see the demand first. These suppliers therefore carry relatively high earnings visibility at this stage of the AI investment cycle.

However, profits will not remain concentrated in hardware forever. As semiconductor supply expands, competition among models intensifies, and inference costs fall, platform companies should gain cheaper access to AI capabilities and weave them into advertising, cloud services, e-commerce, and software of their own.

That is the difference between China Evolution and 3067: the former is built around the picks-and-shovels providers already taking orders, the latter around the next stage of application monetisation — the two strategies capturing different phases of the same AI investment cycle. Picking the right suppliers within that early phase is itself a specialist task, and one particularly well rewarded in China's A-share market, where retail investors generate roughly 70% of turnover with an average holding period of just 13 days. That kind of churn leaves real room for short-term mispricing, which a patient, fundamentals-driven investor can exploit.

That same selectivity, however, cuts both ways. China Evolution's strategy excludes the country's 100 largest companies — a rule that recently left it without exposure to Zhongji Innolight and Eoptolink, two of the strongest performers in China's optical-component rally. Missing out on them dented the fund's relative returns, not because the manager overlooked the trade, but as a structural consequence of the fund's own exclusion rule.

Other holdings in the same optical-communications space helped offset the gap, but the episode is a reminder that excluding mega-caps carries a real opportunity cost alongside its diversification benefit. That trade-off is why China Evolution is best treated as a satellite alongside the core exposure 159939 and 3067 already provide, rather than a replacement for either.

Earnings check: AI spending is reaching the supply chain

The investment case must ultimately be validated by financial results. The latest results from the fund’s ten largest holdings show that stronger demand is already translating into revenue and profits across much of the supply chain.

Table 2: Seven of the fund's ten largest holdings grew profits faster than revenue.

Source: Company filings, Bloomberg Finance L.P. and iFAST Compilations. Portfolio weights as at 31 July 2026; earnings data as at 30 June 2026.

All ten of the fund's largest holdings, as at 31 July 2026, have released H1 2026 results. Across their latest reporting periods, all ten recorded year-on-year revenue growth, while seven delivered headline net profit growth exceeding revenue growth. The holdings span the strategy's three core themes — AI capital expenditure, anti-involution and new consumption — alongside diversified positions outside those themes. Overall, the results indicate that stronger demand is translating into operating leverage across most major holdings.

Jereh’s current financial results require a further distinction. Its power subsidiary has secured substantial orders for gas-turbine generator sets from an international cloud-service provider, including a USD 1.47 billion order scheduled for delivery in 2027. Orders signed with this customer and its subsidiaries since November 2025 have reached USD 1.72 billion. Management has stated that the latest contract will not affect its 2026 results. Jereh’s H1 revenue growth of 10.8% and profit decline of 3.7% therefore largely reflect its existing operations rather than contributions from these data-centre power orders.

The results show that AI-related demand is producing more than additional orders. A richer mix of high-end products, improved pricing and higher capacity utilisation are translating revenue growth into faster earnings growth across most of the major holdings. As earnings become more differentiated, future excess returns should depend increasingly on active stock selection rather than a broad re-rating of the entire hardware sector.

Stock selection has driven long-term excess returns

The fund’s longer-term record remains positive, although its returns have not followed a smooth path. Based on the latest available data as at 27 August 2026, China Evolution gained 30.4% in 2026 year to date, while the MSCI China All Shares Index Net declined by 2.8%, producing an excess return of 33.2 percentage points.

Over the past five years, the fund delivered an annualised return of 3.39%, compared with -0.93% for the index. Since inception, it returned 6.21% annualised, compared with -0.20% for the index.

Chart 1: China Evolution has outperformed across every period shown.

Headline returns alone do not show whether the excess return came from stock selection or simply from the fund’s exposure to mid- and smaller-cap companies. Performance attribution helps distinguish between these two sources.

China Evolution is deliberately constructed to perform differently from the comparison index. It invests beyond China’s 100 largest companies and had an active share of 95.7%. This means that 95.7% of the portfolio differs from the index by holding or weight. It gives active stock selection greater scope to add value, but also increases the possibility of prolonged divergence from index returns.

That divergence was evident between 2022 and 2024, when the fund underperformed the index by 1.5, 3.2 and 3.7 percentage points, respectively. In 2022, stock selection detracted 2.4 percentage points, partly offset by a positive 0.9-percentage-point contribution from market-cap allocation.

In 2023, stock selection detracted 2.3 percentage points, while the underperformance of smaller companies relative to mega-caps reduced returns by a further 0.9 percentage points. In 2024, however, stock selection contributed a positive 4.2 percentage points. This was more than offset by a 7.9-percentage-point drag from the fund’s mid- and smaller-cap bias.

The three-year shortfall therefore included two years of weak stock selection, but by 2024 the principal headwind had shifted from company selection to market-cap allocation.

Over the full attribution period, stock selection has nevertheless remained the strategy’s principal source of excess return. Based on the latest manager attribution available through the end of April 2026, stock selection accounted for approximately 81% of the strategy’s gross attributed excess return, with market-cap allocation accounting for the remainder. From the beginning of 2026 to the end of April, stock selection contributed 19.03 percentage points of the strategy’s 26.04-percentage-point gross excess return.

The attribution therefore indicates that the strategy’s longer-term excess return has primarily reflected company selection rather than a persistent tailwind from smaller companies. However, stock selection detracted in both 2022 and 2023, demonstrating that it does not add value consistently every year. The fund can experience prolonged periods of underperformance when individual selections disappoint or when market returns are concentrated among mega-caps.

Chart 2: China Evolution’s maximum drawdown has been shallower than the comparison index’s since inception, but deeper in 2026 year to date.

The fund’s maximum drawdown since inception was 51.9%, compared with 55.3% for the index. However, in 2026 year to date, its maximum drawdown was deeper at 16.8%, versus 12.7% for the index.

The since-inception comparison does not make the fund defensive: a 51.9% historical drawdown still represents substantial capital risk. Nor does the historical result guarantee that the fund will experience smaller drawdowns than the index in future.

Portfolio role: an active satellite, not a replacement

Within a broader China equity allocation, 159939 and 3067 can continue to provide the core technology exposures. The former captures leading A-share semiconductor and hardware companies, while the latter provides exposure to major internet platforms, cloud computing, models, and application monetisation.

China Evolution adds a differentiated source of return. Its active approach allows the manager to select and size companies based on conviction rather than market capitalisation, while investing beyond the companies that dominate mainstream indices.

The fund also broadens exposure to the power infrastructure supporting AI investment. Among its ten largest holdings, Weichai Power provides data-centre backup-power exposure through already-recognised sales of large-bore diesel generator sets, while Yantai Jereh adds further data-centre power-generation exposure through material orders scheduled mainly for delivery in 2027.

However, China Evolution is not a pure technology, AI, or power-infrastructure fund. The strategy also invests in new-consumption companies and businesses benefiting from supply-side capacity discipline. These exposures provide additional sources of return, but they also mean that the fund’s performance will not depend solely on AI investment.

For investors who already have a diversified China equity allocation, China Evolution could represent a satellite within that existing China allocation. The position would complement, rather than replace, the core technology exposure provided by 159939 and 3067.

Declaration:

For specific disclosure, at the time of publication of this report, IFPL, through its connected and associated entities, and the analyst who produced this report hold a NIL position in the abovementioned securities.

This research report was prepared with the assistance of artificial intelligence tools. iFAST Financial Pte Ltd does not rely exclusively on AI for content generation. The content of this report — including all investment theses, ratings, price targets and conclusions — has been independently reviewed and verified by the research analyst or analysts to ensure accuracy and professional integrity.