Solar Efficiency Standards Take Effect in January. 400GW Has to Be Rebuilt.

A Threshold, Not a Guideline

On 27 June 2026, the Ministry of Industry and Information Technology, the National Development and Reform Commission and the State Administration for Market Regulation issued three mandatory standards for the photovoltaic industry. GB 47834-2026, covering efficiency limits and grades for crystalline silicon modules and inverters, is the one that binds manufacturers. It takes effect on 1 January 2027, leaving roughly six months of transition from publication. Under Article 25 of the Standardisation Law, products below grade 3 may not be produced, sold, imported or supplied. It is the first mandatory efficiency threshold the Chinese PV industry has faced.

The threshold is set in three tiers by technology route. TOPCon modules must reach 24.0, 23.7 and 23.2 per cent for grades 1 to 3; HJT 23.8, 23.5 and 23.2 per cent; BC 24.2, 23.9 and 23.5 per cent. Three further conditions apply across all routes: coupled environmental stress degradation may not exceed 8.6 per cent; bifaciality must be at least 75 per cent for TOPCon, 85 per cent for HJT and 70 per cent for BC; and at the 1134mm by 2382mm format, the grade-3 power band is about 630W for TOPCon and HJT and about 635W for BC.

The number that matters is the gap between them: the BC floor sits 0.3 percentage points above the other two routes. That gap recurs throughout this article, because it is where the repricing of process talent begins.

The standard is easy to misread as industry self-discipline. It changes procurement rules instead. At a Jiangsu PV conference in September, Bai Hualin of the China Photovoltaic Industry Association estimated that about 400GW of domestic capacity faces upgrade or retirement, against assets in the tens of billions of renminbi. Wang Jin of the Guohe Energy Research Institute projects that around 30 per cent of relatively backward capacity will exit once the rules take hold in 2027.

State-owned buyers have already moved. Huadian raised the efficiency floor in its second 2026 module tender to 23.2 per cent, 0.4 points above the first round. Huaneng, Three Gorges and PetroChina now specify grade 2 or grade 1 efficiency in their tenders. Once the buying side moves, the retrofit timetable compresses: two months remain before 1 January, and commissioning a cell line is measured in quarters.

Why BC Sits 0.3 Points Higher

In a laboratory, 0.3 of a percentage point is noise. On a production line it decides how much of a batch falls below the floor, and how quickly an unstable line turns that into scrapped wafers. For manufacturers already running above 24 per cent with stable yields, the threshold is barely a constraint. For those running older lines at volume, 0.3 points is the difference between spending on a retrofit and writing the asset down.

Demand sends a clearer signal than policy. Among the top eleven module makers in the first half of 2026, TOPCon took about 83 per cent of shipments and BC about 16 per cent. BC was roughly 4 per cent in the first half of 2024 and about 6 per cent a year later. LONGi shipped 19.55GW of BC modules, up 125 per cent and above 65 per cent of its mix; Aiko shipped 9.39GW against 8.57GW. Across the top eleven, BC approached 30GW in six months.

The premium is the other half of the story. BC modules currently carry RMB 0.05 to 0.10 per watt more than TOPCon, a margin of roughly 7 to 12 per cent. In an industry that has priced at cost for two years, that spread is what justifies retrofit capital. TCL Zhonghuan plans RMB 2.6 billion of line conversion, with all cell capacity switched to BC by the end of the first quarter of 2027.

The full picture needs stating. BC has not displaced TOPCon, which still holds more than 80 per cent of shipments and which the China Photovoltaic Industry Association expects to retain about 70 per cent over five years. BC occupies the intersection of two markets: high-end distributed generation, where the absence of front busbars is a visual selling point, and mandatory compliance, where its efficiency ceiling happens to clear the line. That narrow position explains why talent prices are rising on a narrow band. Henderson Executive Search sees the same pattern: demand concentrates on a few process steps, not the whole function.

The First Crack in the Earnings

On the evening of 21 October, Aiko Solar published its third-quarter report. Revenue for the nine months reached RMB 14.47 billion, up 24.79 per cent, with a net loss of RMB 709 million. The third quarter alone told a different story: revenue of RMB 5.29 billion, up 67.89 per cent, and net profit of RMB 44.57 million, positive both year on year and quarter on quarter. It is the first single-quarter return to profit among major cell and module manufacturers in this downturn.

The quality of that profit deserves scrutiny. Gross margin for the quarter was 13.31 per cent, up about 6 points year on year and 5 points quarter on quarter. Inventory days fell from 54 in the first quarter to 41 in the third. Operating cash flow for the nine months was RMB 1.95 billion, up 10.45 per cent. Adjusted net profit nonetheless remained negative at RMB 24.58 million, which means the core business has only just crossed the waterline, carried by product mix rather than scale. Contract liabilities rose from RMB 543 million at the start of the year to RMB 889 million, up about 60 per cent, pointing to a fuller order book.

That book is concrete. As of 18 September 2026, Aiko held about 15GW of ABC module orders scheduled into the first half of 2027, with customer intent running ahead of capacity. On the same day the company announced equipment procurement for its Chuzhou base and a 6GW cell retrofit in Tianjin. An earlier RMB 1.665 billion programme converts 5GW of PERC in Yiwu and 6GW of TOPCon in Chuzhou to BC.

Against peers the contrast is sharp. LONGi reported first-half revenue of RMB 27.05 billion, down 17.58 per cent, and a net loss of RMB 3.68 billion. JinkoSolar reported RMB 24.73 billion, down 22.32 per cent, and a loss of RMB 3.08 billion. The industry is still working through overcapacity and price war: tier-one TOPCon quotes sit at RMB 0.70 to 0.72 per watt in the fourth quarter, tier-two below 0.68, with some transactions at 0.65 to 0.69. October production cuts are being implemented, with the industry expected to trim about 20,000 tonnes. One quarter of profit is not an industry reversal, but it marks a position: retrofit and mix move faster than waiting for prices.

Exports Are Falling, Revenue Is Rising

Export data released by the China Photovoltaic Industry Association on 17 September shows an unusual scissors pattern. Total PV product exports reached about USD 17.18 billion in the first half, up 24.3 per cent. Module volume came to 117.7GW, down 0.5 per cent, while module value reached USD 13.2 billion, up 14.4 per cent, the first recovery since the deep adjustment of late 2023. Value growing faster than volume means unit prices are recovering and the old model of trading price for volume is ending.

The category split explains more than the aggregate. Wafer exports reached USD 780 million, up 29.4 per cent. Cell exports reached USD 3.25 billion, up 87.6 per cent, with volume up 36.8 per cent, the fastest-growing category at close to 20 per cent of the total. The driver is the new module capacity overseas: those plants import cells from China and package locally. In plain terms, finished modules are a smaller share of the trade and core components plus overseas assembly a larger one. That is what PV overseas factory building looks like in the customs data.

Regional demand has shifted as sharply. In the first quarter, Chinese module exports to the Middle East rose 470 per cent year on year, storage 620 per cent, with more than 90 per cent of orders written for three years or longer. Over the twelve months to June 2026, exports to Africa reached 23GW, up 53 per cent, level with the Middle East. Europe remains the core market, but its share has fallen to 38 per cent while the Middle East, Africa and Latin America rose to 22, 14 and 11 per cent.

Capacity has moved faster still. In the first half of 2026, Chinese manufacturers held more than 120GW of module capacity outside China, double the level at the end of 2024 and over 65 per cent of global overseas module capacity. Trina Solar’s 5GW plant in Abu Dhabi started production in June, the first gigawatt-scale base in the Gulf; Tongwei’s 10GW wafer and cell project in Saudi Arabia broke ground for output in the first half of 2027. Policy tightened in parallel: export VAT rebates on PV products were abolished on 1 April, and US duties on cells from India, Indonesia and Laos reached combined rates as high as 268.06 per cent.

The Job Map: Two Kinds of PV Engineers

Spread the public PV openings of the past month across a desk and one phrase keeps appearing: at least two years of BC module mass-production experience. Runergy in Yancheng is hiring a module process manager for BC at RMB 20,000 to 27,000 over seventeen months, covering process development and mass-production stabilisation across 0BB interconnection, low-temperature soldering, film lamination, dispensing and lamination, with five years of module process experience of which two must be BC mass-production management. JA Solar in Yangzhou is hiring five process engineers at RMB 10,000 to 15,000, requiring three years in PV and one in BC, with familiarity across wet chemistry steps.

That two-year condition narrows the pool severely. BC only reached volume in 2024, so the number of engineers who have taken a line from trial production through yield ramp is small and concentrated. Cell process windows are also short: wet-chemical patterning, passivation deposition, laser opening and metallisation interlock, and a parameter change propagates along the chain. An engineer fluent on TOPCon has to rebuild an entire set of judgement calls on a BC line. Henderson Executive Search treats this as the central constraint in current PV searches, because widening the salary band does not solve it.

One detail in the pay bands is easily missed. Overseas posts at RMB 12,000 to 16,000 overlap almost exactly with domestic roles at 10,000 to 15,000. The real separation appears at RMB 20,000 to 25,000 in Indonesia and 35,000 to 55,000 in the United States. Ordinary expatriation has stopped being scarce; what is scarce is the person who can enter a high-cost market and keep an entire line running. The value of the PV equipment engineer is shifting from fixing machines to making a line stable inside an unfamiliar supply chain.

A Visa Is Now Part of the Specification

The conditions on a Thailand posting are blunt. Process engineers, shift supervisors, quality leads and equipment engineers are all asked for prior overseas experience and a valid visa. That is a screening condition, not a courtesy: it compresses recruitment from interview, permit and arrival into interview and arrival, separating those who can board a plane from those needing three months of paperwork.

The manufacturing map behind it is legible. Chinese PV plants now run in Thailand, Vietnam, Laos, Indonesia, Cambodia and the United States, and vertically integrated manufacturers hold assets in several of them at once. There is a fixed lag between a plant being finished and a team being in place, and that lag is the real cost of ramping capacity. A new line takes six months to a year from equipment arrival through commissioning to stable output, longer still when the site crosses languages and supply chains.

The external reference shows the width of the gap. The US Energy and Employment Report and the Interstate Renewable Energy Council put US solar employment above 280,000. Meeting 60 to 70GW by the end of 2026 requires about 355,000 workers, a shortfall of some 53,000. Eighty-six per cent of solar employers report difficulty filling openings. The harder layer is competition from outside: storage, data centres and nuclear all recruit from solar, because SCADA operations, commissioning and land development transfer cleanly. Henderson Executive Search encounters this whenever a client insists on sector-specific experience and finds the candidate holds three other offers.

There are plenty of engineers who understand process. The scarce layer is the one that can also manage overseas contractors, work through local certification and grid-connection requirements, and run equipment management when spare parts take months to arrive.

Three Ledgers on the Employer Side

The first is time. Four hundred gigawatts of retrofit demand is compressed before 1 January 2027, while a line passes through five stages from decision to compliance: design, procurement, installation, trial production and ramp. Equipment lead times alone run to months. The only segment a company controls is its own execution, which means the efficiency of its people. An experienced team needs fewer trial runs; a first-time team may lose a production schedule to one failed attempt.

The second is pay. Domestic BC process roles are not yet expensive: RMB 10,000 to 15,000 sits close to TOPCon equivalents, and module process managers reach 20,000 to 27,000. Price moves sharply where the brief combines BC knowledge with team management and customer audit exposure. When the same shortlist is interviewed by three companies in two months, the offer drifts away from the grade of the role. That is the shape of a short-term squeeze, and the part of recruitment cost most often underestimated this quarter. Henderson Executive Search tracks it before salary discussions begin.

The third is organisational. A retrofit reaches beyond engineering into four other functions. Process rewrites SOPs and control plans. Equipment revalidates process windows. Quality rebuilds the criteria for BC-specific defects such as low CTM, EL failures, cold solder joints, bubbles and misalignment. Procurement requalifies materials and sales re-articulates the premium. When a module process manager's job description includes aligning R&D, quality, equipment and supply chain while hosting customer audits, the company is really hiring someone who can hold five lines together. Henderson Executive Search reads that clause as the real specification.

That makes the build-versus-buy decision concrete. Converting people from a TOPCon line is cheaper and preserves institutional knowledge, but asks them to rebuild expertise inside the shortest window. Buying from outside is immediate, except that the whole industry is bidding for the same few hundred people. Over the past two months, Henderson Executive Search has seen clients choose a hybrid: external hires for wet-chemical patterning and metallisation, internal development for peripheral steps, and project advisers to cover judgement gaps during ramp.

What the Numbers Can and Cannot Support

BC market share exists in three parallel definitions, and the differences come from statistical boundaries rather than errors. On a technology-route basis, BC penetration in 2026 is about 2 to 3 per cent, or 15 to 20GW of shipments. On a market-share basis, BC climbs from 6.7 per cent in 2025 toward 11 per cent in 2026. On the shipment mix of the top eleven module makers, BC reached about 16 per cent in the first half. The denominators differ, and the figures cannot be converted into one another. Any citation should name the basis.

The 400GW figure needs the same care. It comes from an estimate offered by an industry expert at a public conference and describes capacity facing upgrade or retirement. There is no official statistical series behind it. The related projection of 30 per cent capacity exit is a research judgement about what happens after 2027, not a recorded outcome.

Talent-gap numbers deserve the most caution. Widely circulated claims of a 200,000 or 600,000 shortfall in Chinese PV employment come from training providers and service firms, and their scope includes installation and operations roles that sit in a different labour pool from manufacturing process engineers. They cannot be added together and should not be used as evidence of hiring difficulty. This article relies on verifiable substitutes instead: the volume and pay ranges of public vacancies, the hard experience thresholds written into job descriptions, and capacity and order schedules disclosed by listed companies.

One final note on scope. The 120GW figure counts capacity Chinese companies have built abroad in the first half, not global module capacity. Pay figures come from public advertisements and describe ranges rather than averages; Henderson Executive Search tests them against actual offers in each search.

FAQ

Is there really a large pool of engineers with two years of BC mass-production experience?
No. BC reached volume only in 2024, and the people who have taken a line through a full ramp sit in a small number of companies. The practical approach is to unbundle the requirement. Among the four critical steps, wet-chemical patterning, passivation deposition and laser opening share process logic with TOPCon, while metallisation differs most. Setting thresholds by step recruits better than a blanket two-year rule.

How long does a mature process engineer need to move from TOPCon to BC?
It depends on the step. Engineers with wet-process and deposition backgrounds usually hold a shift independently within three to six months. Metallisation and laser steps take six to twelve because equipment and materials differ substantially. Pricing a hire on immediate output overstates short-term value; pricing on six-month independence loses candidates during negotiation. A more reliable structure puts ramp support into the offer, such as a BC-experienced adviser for the first quarter.

Why has the pay gap between a module process manager and a marketing manager widened under BC?
Because the brief has expanded. A BC module process manager covers 0BB interconnection, low-temperature soldering, film lamination, dispensing and lamination, while also resolving CTM shortfalls, EL failures, cold joints, bubbles and misalignment, and hosting customer and system audits. The role carries process, quality and customer-facing duties at once. Few people hold all three, so the price drifts above the nominal grade.

What should a smaller manufacturer do when it cannot fund a full retrofit?
Decide first which capacity must comply before 2027, concentrate resources on the lines that can, and prepare the remainder for impairment or conversion. On people, a full BC team is unnecessary: one core process engineer, an external adviser and internally converted staff will usually suffice. The core engineer owns process-window judgement, the adviser supports critical ramp decisions, and converted staff cover daily operation. The structure costs less than a full build, at the price of dependence on two people who need long-term incentives.

How can a company tell whether a PV headhunter actually knows this market?
Three specific questions are enough. Which of the four core BC steps is difficult to staff, and why. How does the process-window difference between 0BB and low-temperature soldering change how a candidate's experience should be discounted. Roughly how long do visa and arrival take for an overseas base. A consultant who answers all three has done the work. One who offers only network and speed will treat the role as general manufacturing. Professionalism in PV recruitment shows up in whether the adviser can judge if a candidate will work six months from now. Henderson Executive Search applies that test to every brief.