China solar: Main-chain prices stabilized in May, but inventories and falling costs still weigh on the profitability inflection point

Overview

This report is Goldman Sachs' monthly tracking of profitability across China's solar industry chain, focusing on supply and demand, inventories, spot prices, input costs, and changes in cash gross margin and EBITDA margin across each segment since May 2026. The conclusion is that prices across the main industry chain are temporarily stable, but weak industry demand, rising inventories, potential upstream supply increases, and falling module production costs may still bring subsequent downside price pressure.

Core views

The core view is that price stability does not mean a profitability inflection point has been firmly established. The module segment has seen implied cash profitability from spot prices improve, supported by easing material prices; however, falling glass and film prices reflect pressure from inventories and raw material prices. Global module demand weakened sharply in April, especially with a notable year-on-year decline in China's installation demand, leaving 4M26 demand progress below the full-year forecast path. At the stock level, the report prefers Maxwell, Hangzhou First, and Longi for their new applications, film price increase potential, energy storage, or cost resilience, while remaining cautious on polysilicon and glass.

Analysis framework

Using a monthly industry-chain tracking framework, the report combines China's implied module demand from installations, China's module exports, and US AC-side implied module demand into global module demand, and then estimates cash gross margin, unit gross profit, and EBITDA margin trends across different segments based on spot prices, input costs, and inventory changes in each sub-sector. Stock target prices mainly use 2027E EV/EBITDA multiples discounted back to 2026E.

Methodology notes

• Profitability trackingChina solar profitability tracker

  Monthly tracking of supply and demand, inventories, spot prices, and input costs

  This framework estimates cash gross margin and EBITDA margin trends for covered companies using supply-demand and inventory dynamics in each sub-sector, together with spot prices and input costs. The report notes that the estimates may differ from actual company results because they do not reflect company-level discounts or premiums, nor do they fully capture operating arrangements such as monthly shutdowns or maintenance.

• Valuation methodEV/EBITDA target price method

  12-month target price

  Maxwell's target price of Rmb278 is based on 22x 2027E EV/EBITDA discounted back to 2026E at an 8.3% cost of equity; Hangzhou First's target price of Rmb21 is based on 13x 2027E EV/EBITDA discounted at a 10.5% cost of equity; Longi's target price of Rmb26 is based on 11x 2027E EV/EBITDA discounted at an 11.3% cost of equity.

• Factor frameworkGS Factor Profile

  Growth, financial returns, valuation multiples, and composite percentile

  Goldman Sachs uses growth, financial returns, valuation multiples, and composite metrics to compare stocks' attributes relative to the market and industry peers; growth is typically based on forward sales, EBITDA, and EPS growth, financial returns are based on ROE, ROCE, and CROCI, and valuation multiples are based on P/E, P/B, dividend-related metrics, EV/EBITDA, EV/FCF, and others.

• M&A assessmentM&A Rank

  Potential acquisition probability score

  Goldman Sachs uses M&A ranks from 1 to 3 to assess the probability that a covered company could become an acquisition target, where 1 indicates high probability, 2 medium probability, and 3 low probability; ranks 1 or 2 are typically reflected in the target price through an M&A factor.

Asset mapping & comparison

Structured mapping from thesis to named assets (strengths, weaknesses, peers, risks).

• China solar industry chain

  Research theme and industry exposure

  Strengths

  Prices across the main industry chain have been broadly stable since May, and the module segment shows signs of improving profitability.

  Weaknesses

  Global and Chinese demand have weakened significantly, trading is subdued, and order visibility is limited.

  Comparison

  4M26 global module demand was down 30% year-on-year, weaker than the FY26 forecast path of -12% year-on-year.

  Risks

  Inventory accumulation, rising upstream supply, lower production costs, and weak price expectations may continue to weigh on profitability.

• Maxwell

  One of the preferred names in coverage

  Strengths

  The report is positive on new application opportunities, with a target price of Rmb278.

  Weaknesses

  If demand for new applications falls short of expectations, it may weaken willingness for pilot capex.

  Comparison

  The target price is based on 22x 2027E EV/EBITDA, with different multiple weightings applied to core business and new application orders.

  Risks

  Cancellation of new application orders, tighter export controls on PV equipment, and slower-than-expected development in the semiconductor industry.

• Hangzhou First

  One of the preferred names in coverage

  Strengths

  The report is positive on PV film price hikes and unit profit expansion potential, with a target price of Rmb21.

  Weaknesses

  If PV installations are below expectations, utilization rates and average selling prices for film may come under pressure.

  Comparison

  The target price is based on 13x 2027E EV/EBITDA discounted back to 2026E.

  Risks

  Installation demand below expectations, slower-than-expected capacity expansion, slower-than-expected development of high-end non-PV businesses, and greater procurement pressure from supply disruptions in raw materials such as resin.

• Longi

  One of the preferred names in coverage

  Strengths

  The report is positive on energy storage potential, EBITDA resilience from falling upstream prices, and medium-cycle upside from BC technology, with a target price of Rmb26.

  Weaknesses

  Profitability improvement is still affected by upstream prices, the rollout of cost-reduction technologies, and the pace of BC development.

  Comparison

  The target price is based on 11x 2027E EV/EBITDA discounted back to 2026E.

  Risks

  If demand exceeds expectations and supply-side policies are strongly enforced, polysilicon and glass price rebounds may exceed expectations; slower-than-expected adoption of cost-reduction technologies; slower-than-expected BC development.

• Daqo ADR/A

  Cautious name in the polysilicon segment

  Weaknesses

  The report remains cautious on the polysilicon segment and notes Daqo ADR/A as Neutral/Sell.

  Comparison

  Compared with Maxwell, Hangzhou First, and Longi, the report does not list it as a preferred direction.

  Risks

  Potential polysilicon supply increases in June may worsen upstream supply-demand dynamics and create downside price risk.

• Tongwei

  Cautious name in the polysilicon segment

  Weaknesses

  The report maintains a Sell view on Tongwei.

  Comparison

  Relative to the preferred names, the report classifies it as a polysilicon exposure requiring caution.

  Risks

  Higher polysilicon supply, falling prices, and industry demand below expectations.

• Flat A/H

  Cautious name in the glass segment

  Weaknesses

  Glass prices have fallen 7% since May, inventories rose to 57 days, and profitability has deteriorated; the report maintains a Sell view on Flat A/H.

  Comparison

  The glass segment is performing weaker than the module segment, with module profitability improving month-to-date while glass profitability declines.

  Risks

  Further inventory accumulation, continued price declines, and insufficient visibility on module orders.

Key data

• Global module demand in April 202633GW, -30% month-on-month, -58% year-on-yearCalculated as the sum of China's implied module demand from installations, China's export volume, and implied demand from US AC-side installations.
• 4M26 global module demand156GW, -30% year-on-yearBelow Goldman Sachs' FY26 forecast path of -12% year-on-year.
• China's implied module demand from installations in April 20269.5GW, -79% year-on-yearMarch was 8.9GW, down 56% year-on-year, and April weakened further.
• 4M26 China installation demand51GW, -51% year-on-yearBelow the FY26 forecast path of -25% year-on-year.
• China module exports in April 202618GW, -16% year-on-year4M26 reached 83GW, down 2% year-on-year; APAC, Africa, Middle East, and EU all weakened sequentially.
• Glass inventory57 days, +8% month-to-dateInventory accumulation is an important reason for falling glass prices and worsening profitability.
• Glass and film prices since MayGlass -7%, film -6%Weaker film prices were also affected by an 11% decline in oil prices month-to-date.
• Change in module profitabilitySpot prices imply cash profitability improved by 2 percentage points month-to-dateMainly driven by easing material prices.
• Change in glass profitabilitySpot prices imply cash profitability declined by 10 percentage points month-to-dateMainly dragged down by falling prices.
• Target pricesMaxwell Rmb278; Hangzhou First Rmb21; Longi Rmb26All are 12-month target prices, based on 2027E EV/EBITDA and discounted back to 2026E.

Impact & implications

The implication for the industry is that short-term price stability reflects more of a temporary balance under subdued trading rather than a full repair in supply and demand. If polysilicon supply increases in June, upstream prices continue to soften, and Tier 1 companies use low-cost metal technologies to reduce module costs, module prices may still come under pressure. From an investment perspective, more attention should be paid to stocks with new applications, film price increases, energy storage, or cost resilience, while avoiding polysilicon and glass segments facing greater supply and inventory pressure.

Risks

• Polysilicon supply may increase in June, worsening upstream supply-demand conditions and triggering downside price pressure.
• Softer upstream prices and adoption of low-cost metal technologies by Tier 1 companies may reduce module production costs and push module selling prices lower.
• Glass inventories have accumulated to 57 days; if demand fails to recover, glass prices and profitability may remain under pressure.
• Global module demand and China's installation demand are significantly weaker than the full-year forecast path, potentially dragging down industry utilization rates and price expectations.
• Film prices have weakened due to falling oil prices; if raw material or demand conditions continue to change, unit profitability may remain volatile.
• At the stock level, risks include weaker-than-expected demand for new applications, export controls, capacity expansion, development of non-PV businesses, adoption of cost-reduction technologies, and BC development.

What to watch

• The release of polysilicon supply in June and changes in upstream supply and demand.
• Whether visibility on module orders, trading activity, and price expectations improve.
• Sequential changes in China's installation demand, module exports, and demand across regions such as APAC, Africa, Middle East, and EU.
• Whether glass inventory days and glass prices continue to decline.
• The progress of low-cost metal technology adoption by Tier 1 companies and its impact on module production costs.
• Maxwell's new application orders, Hangzhou First's film price hikes, and Longi's progress in energy storage and BC technology.