From Scale Advantage to Technology Lead: China’s Battery Supply Chain and the Case for Exposure

Where Earnings Now Come From: Three Drivers and One Divergence

Earnings in this cycle are no longer driven by new energy vehicle sales. Three drivers now dominate: the globalisation of storage demand, share consolidation among leaders under the new national standard, and the transition of next-generation technology from laboratory to production line. Together they point to one conclusion — the earnings recovery is structural, not a broad-based sector upturn. The beneficiaries sit in battery cells and lithium battery materials, while vehicle OEMs and motor control remain weighed down by domestic price competition and the phase-down of the purchase tax exemption.

Within the cell layer, the question of sodium-ion battery share deserves a separate answer. Sodium-ion is displacing lithium iron phosphate in storage and entry-level power applications, but this substitution does not move profit away from the current leaders: industrialisation is being led by CATL and EVE Energy themselves. CATL launched its second-generation sodium-ion product in April 2026 and signed a three-year, 60GWh supply agreement with a storage system integrator, while EVE Energy’s sodium-ion storage systems are already in operation. Sodium-ion is therefore a share tool used by the same companies within the cell layer. It expands the serviceable market and hedges lithium price volatility rather than changing where profit accrues. This is why we do not treat sodium-ion as a separate risk factor in our conclusions.

The Gap Between Price and Earnings: How It Formed

The CSI All-Share New Energy Vehicle Index has fallen steadily through 2026, down roughly 11% year-to-date and around 19% over the past 60 trading days (as at 13 August 2026). Over the same period, earnings at the battery and materials layers improved. This divergence is the starting point for our valuation view, so the mechanism behind it needs explaining rather than simply describing.

First, index weight structure is a direct drag. Within the fund’s top ten holdings, BYD carries an 8.48% weight and Inovance Technology 6.23%. These are the only two companies whose Q1 recurring net profit declined, and their share prices fell in parallel — BYD was down 24.28% over Q2. Roughly 15% of weight therefore contributed negatively on both earnings and price. The cell and materials layers carry higher combined weight, but their price recovery has not yet been sufficient to offset that drag. Index-level weakness is, to a significant degree, a weighted-average outcome.

Second, the sector is being discounted on a mismatched macro narrative. Domestic vehicle sales totalled 9.921 million units in H1 2026, down 21.1% year-on-year, with domestic passenger vehicles down 24.3% — the sharpest fall in domestic demand in recent years. The market has anchored on this figure and applied a discount to the entire new energy vehicle sector. Demand for batteries and materials, however, has already shifted: storage and exports now supply most of the incremental volume, and their outlook is not determined by domestic passenger vehicle sales. In our view this is a case of mispricing — a domestic macro narrative built on autos and property is being applied to a segment that has largely decoupled from it.

Third, earnings confirmation lags. Q1 results were not fully disclosed until late April, and the delivery schedule and earnings contribution from storage orders will only be verifiable in the interim report. Pricing naturally reflects fundamentals with a delay.

The first two mechanisms are the more important. We therefore assess that the valuation gap stems principally from index composition and narrative mismatch, rather than from genuine disagreement about the industry outlook. Gaps of this kind can close once earnings are repeatedly confirmed, though the timing is difficult to predict.

Earnings Divergence: Using the Fund’s Top Ten Holdings

To avoid sample selection and cross-period comparison issues, we use the fund’s top ten holdings directly, applying a single measure throughout: year-on-year growth in Q1 2026 recurring net profit attributable to shareholders — that is, net profit excluding one-off items such as asset disposals and government grants, which better reflects core operating earnings. The table is ordered by portfolio weight.

Source: iFinD, compiled by iFAST Research. Portfolio weights as at 30 June 2026; earnings data are for the single quarter Q1 2026.

CATL, the largest holding at 9.70%, grew 53%, while the second and third largest — BYD and Inovance — declined 49% and 15% respectively. The divergence therefore runs directly through the portfolio’s core weights. Overall, eight of ten grew, representing roughly 36% of weight, against two decliners at roughly 15%. Pressure is concentrated in vehicle OEMs and motor control, while battery cells and upstream materials generally posted strong growth.

A note on base effects. Growth rates at Tianqi Lithium, Do-Fluoride and Tinci Materials reflect an extremely low base in the comparable 2025 period; some of these companies were loss-making or barely profitable at the time, so the multiples are not comparable across companies and are not sustainable. Ganfeng Lithium, also in lithium resources, is in a similar position. Excluding these four, four of the remaining six still grew — Huayou Cobalt +69%, CATL +53%, EVE Energy +36% and Sanhua +16% — for combined weight of roughly 23%, above the roughly 15% carried by the two decliners. More importantly, CATL and EVE Energy, the two largest cell holdings, delivered 53% and 36% growth against a normal earnings base, with no reliance on base effects. The divergence therefore holds after removing low-base names, though its magnitude narrows materially. This is why our earnings forecast does not extrapolate battery-layer growth rates across the index.

Domestic Demand, BYD and Exports: Pull-Forward, Inflection and Offset

The scale of the domestic decline has been greater than we expected at the start of the year. H1 2026 domestic vehicle sales fell 21.1% year-on-year and domestic passenger vehicles fell 24.3%, with domestic new energy vehicle sales also negative. Industry profit fell roughly 20% year-on-year over January to May, with the sales margin down to 3.4%. The principal cause is the change to the new energy vehicle purchase tax incentive: from 1 January 2026 the full exemption became a 50% reduction, and the passenger vehicle cap fell from RMB 30,000 to RMB 15,000. Demand was pulled forward into Q4 2025, leaving a vacuum in Q1 2026.

For BYD, Q1 was the window in which this pull-forward effect was heaviest: recurring net profit fell 49% year-on-year and H1 sales volume fell roughly 37%. Data from the other side of the same company points in the opposite direction. Overseas markets continue to accelerate: May overseas sales reached 161,000 units, up 80.4% year-on-year, with cumulative January–May exports of 614,000 units, up 64.2%. Capacity build-out is progressing in parallel, with the Thailand and Brazil plants in stable volume production and construction underway in Hungary, Turkey and Indonesia. Earnings quality has also improved, with Q1 gross margin recovering sequentially to 18.81%. If the domestic pull-forward effect is indeed fading, as the monthly data suggest, and overseas deliveries continue at the current pace, Q1 is likely to mark a cyclical low for BYD’s earnings. We think this possibility warrants attention, but it should be stated clearly: this view depends on domestic demand not deteriorating further in H2 and on overseas capacity ramping as planned. Until the interim report confirms both, it does not constitute an established inflection point.

At the industry level, exports have become the principal offset to weak domestic demand. H1 2026 vehicle exports reached 5.096 million units, up 65.3% year-on-year, of which new energy vehicle exports were 2.355 million units, up 120%. June alone saw 523,000 new energy vehicle exports, up 160% year-on-year, lifting the export share of sales from 19.7% a year earlier to 33.9%. For the battery layer, exported vehicles and overseas localised capacity together form an installation base that is not directly exposed to domestic retail volatility.

Energy Storage: Demand Is Being Reshaped by Compute

Energy storage is the principal source of incremental battery earnings in this cycle. According to the China Automotive Battery Innovation Alliance, combined power and storage battery sales reached 783.4GWh over January to May 2026, up 48.5% year-on-year. Storage batteries accounted for 255.5GWh of that, up 87.7% — close to three times the growth rate of power batteries.

More significant is that the driver of this increment is now partly independent of new energy vehicles. Data centre requirements for power stability and build speed have turned storage from an optional configuration into a standard one. CATL’s Hong Kong listing prospectus discloses that data centre storage battery shipments are forecast to grow from approximately 10GWh in 2024 to around 300GWh by 2030, a CAGR of roughly 76%. As at end-2025, China had approximately 136GW/351GWh of new-type storage in operation, up 84% year-on-year. On policy, compute–power coordination — the integrated planning of computing facilities alongside power systems — was written into the Government Work Report for the first time in 2026 and incorporated into the 15th Five-Year Plan outline, requiring computing facilities to be planned together with new-type grids and green power.

What this means for the fund is specific: CATL at 9.70% of NAV and EVE Energy at 4.90% both run storage businesses serving two demand sources — domestic computing infrastructure build-out and overseas data centre storage — neither of which depends on domestic passenger vehicle penetration. This is the core basis for our view that battery-layer earnings can separate from the vehicle OEMs.

Solid-State Batteries: Optionality Outside the Base Case

Solid-state batteries entered pilot in-vehicle validation in 2026 (pilot production being the engineering stage between laboratory and mass production), with mainstream OEMs planning small-volume vehicle installation in 2027 and sulphide the preferred technology route among leading players. The industrial significance is a change in how these names are assessed — validation moves from the laboratory to OEM production lines, and the basis for pricing shifts from technology concept to deliverable volume production capability.

It should be stated clearly that solid-state batteries are not included in our earnings forecast or target valuation. The EPS path set out below is derived entirely from profit growth at the cell and materials layers, exports and storage, and the 22x fair P/E assumption incorporates no valuation expansion. In this report, solid-state batteries are positioned as optionality for 2027 and beyond: if in-vehicle validation proceeds smoothly, the effect would first appear as an improvement in sector risk appetite and the valuation anchor, rather than as an earnings contribution within the 2026–2028 base case. We exclude it from the calculation so that the target level rests only on earnings sources already confirmed in reported results.

Risk Factors

•    Slower-than-expected domestic demand recovery: H1 domestic vehicle sales fell 21.1% year-on-year. If the purchase tax pull-forward effect fades more slowly than expected, or price competition persists, the earnings drag from vehicle OEMs and motor control could exceed our assumptions and affect the index EPS path.

•    Upstream low-base effects are not sustainable: High growth rates in lithium resources and parts of the materials chain rest on an extremely low 2025 base. As the base normalises, year-on-year growth in these segments will fall materially. A renewed decline in lithium prices would further limit the scale of earnings improvement.

•    Supply-chain correlation risk: Within the top ten holdings, lithium resources and battery cells sit on the same supply chain. Sharp moves in lithium prices, or changes in domestic new energy vehicle policy, could cause a high proportion of weight to move in the same direction, materially weakening the diversification benefit under such shocks.

•    Technology route and commercialisation timing: The solid electrolyte route has not yet converged, and the 2027 mass-production installation schedule could slip. In an all-solid-state system, solid electrolytes also pose long-term substitution pressure on conventional separators.

•    Overseas trade environment: Exports have become the principal offset to weak domestic demand. Further tightening of tariff or non-tariff barriers in major export markets would directly affect export growth and overseas capacity utilisation.

Using an Index Vehicle: What Is Actually Being Diversified

At a stage when earnings diverge sharply across segments and the next-generation technology route has not yet converged, single-stock selection carries two risks at once: choosing the wrong segment and choosing the wrong route. Three solid electrolyte paths — oxide, sulphide and polymer — remain in parallel development, and the application boundary between sodium-ion and lithium-ion is still being drawn. Backing a single company is, to a significant degree, equivalent to backing a technology route that has not yet converged.

We need to qualify what diversification means here. Within the top ten holdings, lithium resources (Tianqi, Ganfeng) and battery cells (CATL, EVE Energy) sit on the same supply chain, so a sharp move in lithium prices or a change in domestic new energy vehicle policy could cause roughly 51% of combined weight to move in the same direction. What the fund provides is therefore not full diversification of supply-chain risk, but diversification of end-market exposure — the portfolio spans domestic and export demand, power and storage, and vehicles and materials. With domestic passenger vehicles under pressure while storage and exports grow strongly, this layer of diversification is meaningful in practice. H1 2026 performance across the top ten illustrates the point: BYD fell 24.28% and Ganfeng Lithium 16.56% over the period, while Do-Fluoride rose 81.48%, Tinci Materials 12.43% and Tianqi Lithium 10.53%.

Valuation and Three-Year Return Potential

We build a three-year earnings framework using constituent-weighted EPS for the CSI All-Share New Energy Vehicle Index. The 2026 growth assumption of 35% is a weighted result of the divergence described above. On the upside: actual results at the cell and materials layers, storage sales up 87.7% year-on-year, and exports up 120%. Constraining that: the vehicle OEM and motor control names, roughly 15% of top-ten weight, and the actual decline in domestic demand. The assumption already accounts for the unsustainability of upstream low-base effects and does not extrapolate battery-layer growth rates across the index. Growth then eases to 30% and 25% in 2027 and 2028 as the base rises.

_{Source: iFinD, compiled by iFAST Research; E denotes forecast. P/E in the table is the latest index level divided by each year’s EPS, showing the valuation the current price implies against each year’s earnings. Because H1 2026 earnings have already grown materially, the trailing P/E shown on market terminals is calculated on a different basis from the 2025 figure above. The target level is a scenario calculation based on a 22x fair P/E and forecast 2028 EPS; potential upside is measured against the 12 August 2026 close of 3,959 points.}

The 22x fair P/E sits at the 15th percentile of the index’s P/E distribution since 2012, within the historically undervalued range. We do not assume any re-rating toward the historical median, so the target upside is driven almost entirely by earnings growth. Applying 22x to forecast 2028 EPS of RMB 4.23 gives a target index level of approximately 5,175 points, implying roughly 31% three-year potential upside from the 12 August 2026 close of 3,959. This upside moves with the index level: the target is set by the earnings forecast and the fair P/E, so the implied potential has narrowed following the index’s rebound in recent months. As noted above, the calculation excludes any valuation uplift from solid-state battery commercialisation, which sits outside the base case.

Recommended Vehicle

On the basis of the above, we would highlight the ChinaAMC CSI New Energy Vehicle ETF (515030.SH). The fund is managed by China Asset Management and has been run by Li Jun continuously since inception in February 2020. It tracks the CSI All-Share New Energy Vehicle Index and, as at mid-July 2026, had delivered a cumulative return of 55.02% since inception, 7.57 percentage points above its benchmark, operating at close to full investment. Its portfolio relevance rests on three points: power batteries and materials together account for more than 40% of index weight, matching the segments benefiting most from storage and export demand in this cycle; the portfolio spans domestic and export demand as well as power and storage, avoiding concentrated exposure to a single direction at a time when earnings diverge across segments; and the largest cell holdings are themselves leading the industrialisation of sodium-ion and solid-state technology, so progress in next-generation technology can flow directly into constituent earnings and valuations. We would note that the views in this report correspond to an investment horizon of around three years; the fund carries high concentration and volatility and is not suited to a short-term trading approach.

_{Source: Fund manager and index provider (public disclosures), compiled by iFAST Research.}

Declaration:

This research report was prepared with the assistance of artificial intelligence (AI) 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(s) to ensure accuracy and professional integrity.

For specific disclosure, at the time of publication of this report, IFPL (via its connected and associated entities) holds a NIL position in the abovementioned securities. The analyst who produced this report holds positions in iShares Hang Seng TECH ETF and Global X Asia Semiconductor ETF.