Canadian Solar Inc. (NASDAQ: CSIQ) has reported a sharply divided second quarter in which battery energy-storage shipments increased 73% year over year to 3.7 GWh and exceeded the company’s 2.8-3.2 GWh guidance, while solar-module shipments fell 60% to 3.1 GW and total revenue declined 29% to US$1.208 billion. Gross profit fell to US$168.5 million from US$505 million a year earlier, compressing gross margin to 13.9% from 29.8%, while Canadian Solar recorded a US$77 million attributable net loss compared with US$7 million of net income in Q2 2025. The quarter increasingly separates the company into two narratives: difficult global solar manufacturing economics and a storage operation whose shipment volumes, backlog and project pipeline continue expanding rapidly.
Canadian Solar also ended June with approximately US$1.9 billion of cash including restricted cash and total debt and financing liabilities of US$7.1 billion, up from US$6.8 billion three months earlier. Approximately US$4.1 billion of debt relates to Recurrent Energy, US$2.5 billion to manufacturing and US$0.4 billion to convertible notes, while US$2.6 billion of Recurrent Energy borrowing is non-recourse project debt. Operating activities consumed US$181 million of cash during the quarter, keeping asset sales, project financing and disciplined capital allocation central to the company’s ability to fund growth.
The strategic response is an increasingly U.S.-focused manufacturing and storage buildout. Canadian Solar has opened the first 2.1 GW phase of its heterojunction solar-cell factory in Jeffersonville, Indiana, plans to increase U.S. cell nameplate capacity to 6.3 GW during the first half of 2027 and is expanding its Texas module facility from 5 GW to 10 GW. Those investments could create a more valuable domestic manufacturing position, but management cautions that ramp-up costs at the Indiana plant will continue weighing on profitability through the remainder of 2026.
Why is the 3.7 GWh storage result the strongest part of Canadian Solar’s quarter?
Battery energy-storage shipments increased 82% sequentially from Q1 and 73% year over year, reaching 3.7 GWh against management guidance of only 2.8-3.2 GWh. The result exceeded the high end of guidance by approximately 0.5 GWh, or about 15.6%, even though 471 MWh of those shipments went to internal projects under execution and will have their associated revenue recognized in subsequent quarters.
Storage generated US$425.9 million of manufacturing revenue in Q2 compared with US$589.4 million from solar modules. That means battery systems already produced revenue equal to roughly 72% of module revenue during the quarter, a striking change for a company historically identified primarily with photovoltaic manufacturing. Storage revenue was slightly below the US$432.4 million generated a year earlier despite much higher shipment volume, indicating that pricing and mix have changed materially even as physical deployment accelerated.
The more important visibility metric is e-STORAGE’s US$3.5 billion contracted backlog, which includes signed customer commitments and long-term service agreements. Unlike the much larger early-stage development pipeline, management describes the e-STORAGE backlog as binding, giving the company a clearer base of future revenue while solar manufacturing remains exposed to severe price competition.
How much weaker has Canadian Solar’s solar-module business become?
Q2 module shipments recognized as revenue were 3.1 GW, down 60% from the prior-year period despite increasing 25% sequentially. Module revenue fell to US$589.4 million from approximately US$1.022 billion a year earlier, a decline of about 42%, demonstrating that weaker shipment volume rather than price alone drove the year-over-year contraction.
The difference between the 60% shipment decline and 42% revenue decline also suggests average revenue per watt improved relative to the prior-year mix, although geographic mix, product specifications and customer contracts make a simple price calculation imprecise. Nearly half of Q2 module shipments went to North America, where Canadian Solar is deliberately prioritizing higher-value domestic supply rather than maximizing worldwide shipment volume.
Management has consequently stopped emphasizing aggregate global manufacturing capacity and is focusing disclosure on strategic markets. That shift reflects the reality that enormous manufacturing volume is not necessarily valuable when global module oversupply drives prices toward or below economic production levels.
Why did gross margin collapse from 29.8% to 13.9%?
The year-earlier comparison benefited from release of unrealized profit associated with a U.S. project accounted for as a sales-type lease, while the first quarter of 2026 benefited from an IEEPA tariff refund. Neither benefit repeated at the same level in Q2, causing reported gross margin to normalize sharply even before considering underlying product economics. Gross profit consequently fell approximately 67% year over year to US$168 million.
Storage margins also normalized during the quarter, while Canadian Solar continues dealing with lithium carbonate pricing and competitive solar markets. Operating expenses of US$240 million exceeded quarterly gross profit by approximately US$71 million, producing a corresponding operating loss of about US$71.1 million. The relationship shows why a 13.9% gross margin is currently insufficient to cover the group’s manufacturing ramp-up, logistics, project-development and corporate cost base.
Canadian Solar expects Q3 gross margin between 13.5% and 15.5%, indicating that management does not anticipate a rapid return to the unusually high margins reported in Q2 2025 or Q1 2026. Revenue is expected to improve sequentially to US$1.3 billion-US$1.5 billion, but the quality of that increase will depend on whether greater shipment volume generates enough gross profit to offset ongoing U.S. manufacturing ramp costs.
How much larger could U.S. manufacturing become by 2027?
Canadian Solar’s Jeffersonville HJT cell factory currently has 2.1 GW of Phase I nameplate capacity, with another 4.2 GW planned through Phase II. Trial production for the second phase is expected during the first quarter of 2027, bringing total U.S. solar-cell nameplate capacity to approximately 6.3 GW in the first half of the year.
The Mesquite, Texas, module plant is simultaneously being expanded from 5 GW to 10 GW of annual nameplate capacity. Combining 6.3 GW of planned U.S. cell manufacturing with 10 GW of domestic module assembly would give Canadian Solar a considerably more integrated North American supply chain, although the difference in cell and module capacities means the Texas plant will still require additional cell supply beyond Jeffersonville if operated near maximum output.
Domestic production can potentially capture U.S. incentives, reduce exposure to tariffs and import restrictions and satisfy customers seeking locally manufactured content. The cost arrives first, however, because management expects Jeffersonville ramp-up expenses to pressure earnings throughout the remainder of 2026 before higher utilization can improve unit economics.
Why does Canadian Solar carry $7.1bn of debt?
A large portion of the debt finances power and storage projects rather than simply covering corporate operating losses. Recurrent Energy carried approximately US$4.1 billion of debt and financing liabilities at June 30, including US$2.6 billion of non-recourse borrowings linked to specific projects, while manufacturing carried approximately US$2.5 billion. Non-recourse debt generally limits creditor recourse primarily to the underlying project assets and cash flows, which makes it economically different from holding the same amount entirely as parent-company unsecured debt.
The absolute figure remains substantial. Total debt increased approximately US$300 million during Q2 as Recurrent drew additional project financing for U.S. solar and battery developments, while quarterly operations consumed US$181 million of cash. Canadian Solar therefore needs project completions, asset sales and operating cash generation to recycle capital rather than allowing debt growth to indefinitely outrun monetization.
Interest expense reached approximately US$63.6 million during Q2 and US$116 million in the first half. Recurrent Energy accounted for about US$41.9 million of the quarterly total, reinforcing the importance of project sale timing and electricity revenue in determining whether the development portfolio earns returns comfortably above its financing cost.
What happened to Recurrent Energy during the quarter?
Recurrent Energy generated only US$117.3 million of Q2 revenue and an operating loss of approximately US$19.4 million as planned project sales slipped into the second half. Management described the quarter as light rather than suggesting the assets had been cancelled, and expects delayed sales to contribute to a stronger Q3. Electricity revenue also increased sequentially following commercial operation of a major Spanish utility-scale solar project.
The project-development business remains enormous relative to quarterly revenue. Canadian Solar reports a 21.7 GW solar pipeline and 84.1 GWh of battery-storage projects, including 1.7 GW of solar already under construction and 600 MWh of battery projects under construction. Only 2.2 GW of the solar pipeline and 4.4 GWh of the storage pipeline are classified as backlog, while most of the remainder remains in advanced or early-stage development.
That distinction is crucial because early-stage project pipelines are not contracted revenue. Canadian Solar itself warns that total pipeline size does not predict future owned generation, storage assets or financial results because projects can be sold, delayed, transferred or fail to progress through development.
Why does the 84.1 GWh storage pipeline need careful interpretation?
The headline pipeline is geographically diversified, with approximately 32.3 GWh in EMEA, 23 GWh in North America, 16.6 GWh in Asia-Pacific and 12.2 GWh in Latin America. Yet approximately 71.2 GWh, or nearly 85% of the entire pipeline, remains in early-stage development, while only 600 MWh is under construction and approximately 4.4 GWh is in backlog.
That means investors should not value all 84.1 GWh as though construction or revenue were guaranteed. The more reliable commercial indicator is e-STORAGE’s separate US$3.5 billion contracted equipment and service backlog, because those orders represent binding customer commitments rather than potential development opportunities.
The distinction also demonstrates the two different ways Canadian Solar participates in energy storage. It manufactures and sells battery systems through e-STORAGE while Recurrent develops projects that may eventually be built, sold or retained as operating assets, giving the group opportunities across both equipment supply and project ownership but exposing it to different capital and execution risks.
Can Q3 improve enough to change the profitability picture?
Management expects Q3 revenue of US$1.3 billion-US$1.5 billion, with 3.5-3.8 GW of module shipments and 3.4-3.8 GWh of storage shipments. At the midpoint, revenue would rise to approximately US$1.4 billion, about 16% above Q2, while module shipments would increase approximately 18% and storage volumes would remain close to the exceptionally strong second-quarter level.
Gross margin guidance of 13.5%-15.5% suggests the company expects stability rather than a dramatic profitability rebound. Applying the 14.5% midpoint mechanically to US$1.4 billion of revenue would produce about US$203 million of gross profit, roughly US$35 million above Q2 but still requiring strict control of operating expenses to restore positive operating income. That illustrative calculation is not company guidance for gross profit because actual revenue and margin mix can vary within the disclosed ranges.
Canadian Solar is also maintaining its U.S. 2026 shipment targets of 6.5-7 GW of modules and 4.5-5.5 GWh of battery storage, demonstrating that the domestic manufacturing and energy-storage strategy remains intact despite the Q2 loss. The real second-half test will be whether higher-value U.S. shipments and deferred Recurrent asset sales improve cash generation enough to justify continued expansion.
What is the central investment tension after Canadian Solar’s Q2 result?
The strongest growth engine is increasingly storage rather than conventional modules. Storage shipments rose 73%, e-STORAGE has US$3.5 billion of binding backlog and Canadian Solar is guiding another 3.4-3.8 GWh for Q3, while module shipments remain dramatically below their year-earlier level.
At the same time, Canadian Solar is financing a major U.S. manufacturing expansion and an enormous global development portfolio while carrying US$7.1 billion of debt and consuming operating cash. Those investments could create valuable domestic manufacturing, contracted storage revenue and recurring electricity income, but they require successful project sales and utilization before the balance-sheet burden becomes less prominent.
The Q2 numbers therefore do not fit a simple bullish or bearish description. Canadian Solar is losing money while one of its newest businesses is scaling faster than guidance, and it is building a large U.S. industrial position precisely when global solar manufacturing economics remain weak. Whether that divergence ultimately creates value depends on storage margins, U.S. factory economics and Recurrent Energy’s ability to turn projects into cash faster than debt continues to grow.
Discover more from Business-News-Today.com
Subscribe to get the latest posts sent to your email.
