Prolepsia forecastYes / no

Will China's newly installed solar power capacity in 2027 be higher than in 2026, as reported by China's National Energy Administration?

As of 17 November 2026, Prolepsia puts the chance of yes at 74%.

74%chance of yes

Yes 74%No 26%
Prolepsia forecast made

How Prolepsia reasons

The report Prolepsia wrote with this forecast.

How forecasts are made · Track record

TL;DR

The forecast is 74.4% probability of Yes and 25.6% probability of No. A rebound in China’s annual grid-connected PV additions is favored because 2026 is a depressed comparison year and major project-completion deadlines support activity in 2027; a return to the earlier installation peak is not required.

Key evidence

  • The comparison base is low. The latest NEA-attributed table reports 110.329 GW of PV additions during January–September 2026, against 239.528 GW during the corresponding period of 2025 (2026 table, 2025 table). These are partial-year figures, not the final annual totals.
  • There is a specific commissioning incentive. The NEA aims for substantial completion of the second and third batches of predominantly locally consumed wind-and-solar bases by end-2027. This supports recovery, but the announcement does not quantify the remaining PV-only capacity or its commissioning year (NEA statement).
  • Industry expectations are genuinely divided. Sungrow expects Chinese installations to recover in 2027, citing improving provincial tariffs and infrastructure construction. JA Solar expects slightly declining domestic demand, pointing to connection limits, weak midday revenues and slower approvals (Sungrow record, JA Solar record).
  • Absorption remains a constraint. National PV utilization improved from 91.4% in the first half of 2026 to 93.1% over January–September. Those cumulative readings show improvement, not the disappearance of integration problems (half-year briefing, nine-month briefing).

Uncertainties

The largest unknown is the final quarter of 2026. Strong commissioning would lift the annual threshold and could bring forward projects otherwise expected in 2027. Persistent revenue pressure, distributed-grid bottlenecks or delayed base projects could prevent recovery altogether.

Resolution depends on comparable NEA-reported annual PV additions, including centralized and distributed installations—not cumulative capacity or electricity generation. Equality resolves No. Official revisions available by June 30, 2029, can affect the comparison; insufficient comparable official data would lead to annulment rather than substitution of industry estimates.

Read the full report

TL;DR

The forecast is 74.4% probability of Yes and 25.6% probability of No for China’s NEA-reported annual grid-connected PV additions in 2027 exceeding 2026. The central case is recovery from a depressed comparison year, supported by project-completion deadlines rather than a return to record installation levels. The decisive risks are a strong late-2026 commissioning surge and persistent barriers to profitable grid connection.

Context

As of November 17, 2026, the first comparison year is unfinished. The latest NEA-attributed table records 110.329 GW of newly connected PV during January–September 2026, including both centralized and distributed systems (nine-month release). That establishes a weak starting point for the annual comparison, but it does not establish the full-year threshold that 2027 must exceed.

The question concerns annual newly grid-connected photovoltaic capacity in mainland China. It does not concern cumulative solar capacity, manufacturing output or electricity generation. Direct annual NEA additions take precedence over changes in cumulative capacity, and the latest comparable official revisions available by June 30, 2029, govern resolution. Equal reported annual values resolve No.

Evidence

The historical backbone favors growth, but it also shows why a rebound is not automatic. The historical series covering 2015–2025 contains eight increases in ten annual comparisons. That record has limited relevance to today’s much larger market and its changed electricity-pricing rules (historical series). The earlier policy downturn included seven declining quarters out of eight, showing that adjustment can extend beyond a single year (historical downturn analysis).

The annual observations explicitly identified in the evidence illustrate both the earlier contraction and the later expansion:

Calendar year Newly installed PV capacity
2018 44.26 GW
2019 30.11 GW
2024 277.570 GW
2025 316.574 GW

The earlier annual figures and downturn are described in the historical research and NEA reporting (2018 NEA release, downturn review). The later figures come from the detailed PV construction statistics (2024 table, 2025 table). These observations are not a continuous annual history and should not be read as one.

The strongest current evidence is the depressed comparison base. January–September PV additions fell from 239.528 GW in 2025 to 110.329 GW in 2026 (2025 nine-month table, 2026 nine-month table). The detailed first-half 2026 total was 71.768 GW; subtracting it from the nine-month figure gives an indicative third-quarter addition of 38.561 GW. That subtraction spans publication vintages and is not a separately reported quarterly statistic (half-year table, nine-month release). I read this as stabilization after a sharp slowdown, not proof of a broad recovery.

Year-end seasonality prevents a simple extrapolation. Across 2022–2024, roughly 40%–42% of annual additions occurred in the fourth quarter. For 2025, the directly reported annual and nine-month PV figures imply 77.046 GW in the final quarter (seasonal-history reference, 2025 annual table, 2025 nine-month table). Late-year commissioning therefore remains central to the comparison.

The best timing-specific support for 2027 is the NEA’s objective to substantially complete the second and third batches of predominantly locally consumed wind-and-solar bases by end-2027. This is a dated execution incentive, not merely a distant capacity goal. But the announcement supplies neither a remaining PV-only total nor an allocation between late 2026 and 2027 (NEA construction statement). A concrete example is the 3 GW Jiuduntan PV project within the Gansu-to-Zhejiang infrastructure package, whose major projects were scheduled to enter service progressively before June 2027. Some associated capacity can still arrive during 2026 (NEA project inspection).

Policy is responding to the slowdown. The NEA’s development meeting called for organizing 2027 incremental-project pricing auctions and addressing transmission and distributed-generation connection bottlenecks (meeting statement). Provincial auction results also show selective revenue stabilization. Reported solar clearing prices rose from RMB0.1954 to RMB0.2447 per kWh in Gansu and from RMB0.150 to RMB0.259 per kWh in Xinjiang. These are provincial cohort results, not a national tariff floor; eligibility can include already commissioned projects (Gansu comparison, Gansu auction research, Xinjiang comparison).

The economic constraint is real. Document 136 distinguishes projects around the June 1, 2025 commissioning boundary and exposes newer projects to market pricing, with difference settlement covering selected electricity volumes. Higher supported prices do not protect every unit of output or guarantee attractive returns (pricing reform).

Physical absorption is also mixed. PV utilization rose from 91.4% in the first half of 2026 to 93.1% over January–September. These are cumulative absorption measures, not capacity factors or standalone quarterly readings (half-year briefing, nine-month briefing). Operating new-type storage reached 167 GW/435 GWh by September-end, alongside more than RMB600 billion of grid-enterprise investment during the first nine months. Those investments support integration, but they cannot be translated directly into additional PV connections (NEA infrastructure briefing).

Forward-looking industry evidence remains divided. The IEA expects gradual recovery after the 2026 decline, but its solar capacity conventions are not interchangeable with the resolving NEA metric (IEA outlook). Sungrow expects domestic recovery, citing provincial tariffs and base-and-transmission construction (investor record). JA Solar expects slightly declining domestic demand and identifies connection limits, weak midday revenues and slower approvals (investor record). That contrary guidance is a substantive uncertainty, not something to dismiss because the comparison base is low.

What's non-obvious

A recovery can begin before it helps the Yes outcome. Strong late-2026 commissioning raises the denominator that 2027 must beat. It can also pull scheduled projects out of the following year. Better industry conditions therefore support the comparison only when their effect on 2027 exceeds their effect on the unfinished comparison year.

The other distinction is between reported utilization and project profitability. Under the pricing rules, electricity not delivered because of bidding or pricing is excluded from renewable-utilization assessment. An improving utilization reading therefore does not establish a matching improvement in developer revenues (pricing reform). Likewise, higher auction prices do not establish more newly connected capacity. These distinctions explain why policy support and continuing investment can coexist with cautious company guidance.

Uncertainties

The largest gap is the unfinished 2026 annual total. The final quarter will determine the threshold, and commissioning dates will determine whether current project activity benefits 2026 or 2027. Full-year NEA PV additions will close that gap more directly than cumulative-capacity changes.

The second gap is executable, profitable PV pipeline. The evidence does not establish a reconciled national PV-only schedule showing remaining capacity, revenue coverage, connection readiness and commissioning year. Such a schedule would clarify how much of the base-project deadline becomes recognized additions rather than delayed construction.

Finally, statistical comparability matters in a close result. The latest nine-month PV table notes provincial statistical corrections (NEA-attributed table). Broader solar figures also include solar thermal capacity: the 2025 headline of approximately 318 GW differs from the detailed PV total, just as the first-half 2026 all-solar figure of 72.07 GW differs from PV-only additions (annual breakdown, half-year breakdown). Resolution must use comparable PV figures, including official revisions—not mixed technologies, reporting conventions or cumulative totals.

Forecast history

How Prolepsia’s forecast has movedOne forecast so far: 74% on 17 Nov 2026.
DateForecast
74% chance of yes

Sources Prolepsia read

  1. 国家能源局公布前三季度光伏并网数据--中国能源新闻网cpnn.com.cn
  2. 2015-2023 年我国光伏新增装机情况static.cninfo.com.cn
  3. Solar installations in China start to recover from the low pointwww.bernreuter.com
  4. 2018年可再生能源并网运行情况介绍---国家能源局www.nea.gov.cn
  5. 甘肃省新能源增量项目第二批竞量竞价结果已完成公示_国家能源局甘肃监管办公室gsb.nea.gov.cn
  6. Gansu auction researchpdf.dfcfw.com
  7. 新疆2027年上半年机制电价竞价结果解读:光伏电价上涨 72.67% 机制电量同步收窄-索比光伏网news.solarbe.com
  8. 〖关于深化新能源上网电价市场化改革 促进新能源高质量发展的通知(发改价格〔2025〕136号)〗-国家发展和改革委员会www.ndrc.gov.cn
  9. Renewable electricity - Renewables 2026 - IEAwww.iea.org
  10. 新疆2026机制电价竞价结果:风电0.21元/kWh,光伏0.15元/kWh,电量198亿kWh|新疆_新浪财经_新浪网finance.sina.com.cn

The question

This question asks whether China's total newly installed solar power generation capacity during calendar year 2027 (January 1–December 31, 2027) will exceed the corresponding total for calendar year 2026 (January 1–December 31, 2026), according to official statistics published by China's National Energy Administration (NEA). The comparison concerns annual additions of solar photovoltaic (PV) electricity generation capacity, measured in gigawatts (GW), rather than cumulative installed capacity, electricity generated, or the percentage growth rate of solar installations. Both utility-scale and distributed solar installations are included, consistent with the NEA's reporting methodology. China has experienced exceptionally rapid solar capacity expansion in recent years. According to the NEA, China added approximately 318 GW of grid-connected solar generation capacity in 2025, setting a new annual record. This followed approximately 277 GW of new solar installations in 2024. However, the pace of installations slowed in 2026. According to the NEA's July 2026 briefing, China added approximately 72.07 GW of solar generation capacity during the first half of 2026. By the end of September 2026, China's cumulative installed solar generation capacity had reached approximately 1,310 GW, representing year-over-year growth of 16.5%. These developments provide context for assessing whether annual solar capacity additions will rebound in 2027 or continue to decline. Sources: NEA's February 2026 renewable energy statistics (https://www.nea.gov.cn/20260212/7778cf9907bf4eabb0e73a939ff2ade2/c.html), July 2026 press conference (https://www.nea.gov.cn/20260730/3ce671c387574eeeb120fc3825be0399/c.html), and October 2026 reporting (https://xbj.nea.gov.cn/dtyw/hyxx/202610/t20261030_311677.html). The question is being formulated on November 17, 2026, before the end of the first comparison year. Its outcome will be determined after the NEA publishes nationwide annual solar installation statistics for both 2026 and 2027, normally expected during the following calendar years.

How it resolves

The question resolves YES if the National Energy Administration of China reports that the total amount of newly installed solar photovoltaic electricity generation capacity nationwide during calendar year 2027 is strictly greater than the corresponding amount reported for calendar year 2026. The question resolves NO if the NEA reports that newly installed solar photovoltaic generation capacity during 2027 is less than or equal to the amount reported for 2026. Equal values, at the precision published by the NEA, resolve NO. The primary resolution source will be the NEA's official annual renewable energy development statistics and announcements concerning the nationwide grid-connected operation of renewable energy (全国可再生能源并网运行情况) for 2026 and 2027, published through the NEA's official website (https://www.nea.gov.cn/). The relevant metric is annual newly added grid-connected solar photovoltaic generation capacity (光伏发电新增并网装机容量), including both centralized and distributed installations. Where possible, the comparison will use the NEA's directly reported full-year solar capacity additions rather than values calculated from differences in cumulative capacity. This avoids discrepancies caused by capacity retirements, statistical adjustments, or revisions to previously reported cumulative totals. If the NEA publishes revised annual installation figures, the most recent comparable official figures available by June 30, 2029, will be used. If the NEA changes its reporting methodology between 2026 and 2027, officially revised figures or comparable historical series published by the NEA will take precedence over figures calculated using inconsistent methodologies. If annual installation figures are not published directly but can be unambiguously derived by summing the NEA's officially reported monthly additions for each calendar year, those figures may be used, provided they represent the same metric and geographical coverage. If the NEA has not published sufficient comparable information to determine the outcome by June 30, 2029, the question will be annulled rather than resolved using estimates from other organizations. The question will also be annulled if methodological changes make the two annual figures fundamentally incomparable and no official reconciliation is available by that deadline.

Fine print

The geographical scope is mainland China, following the NEA's national electricity statistics. Hong Kong, Macau, and Taiwan are excluded unless the NEA explicitly incorporates them into its reported national totals. The comparison concerns newly installed solar photovoltaic electricity generation capacity, not solar thermal heating installations, solar manufacturing capacity, annual electricity output, or cumulative solar generation capacity. Capacity will be compared in gigawatts (GW), with official figures published in kilowatts or megawatts converted to GW where necessary. A change in the units used for reporting does not affect the outcome. The comparison uses calendar years rather than fiscal years or rolling twelve-month periods. Installations will be attributed to the calendar year in which they are recognized in the NEA's official statistics, regardless of when physical construction began. The metric includes utility-scale and distributed solar photovoltaic installations, including residential, commercial, and industrial rooftop systems, to the extent they are covered by the NEA's nationwide statistics. Projects under construction or awaiting grid connection are excluded until recognized as newly connected capacity under the NEA's methodology. If multiple official NEA publications provide slightly different figures, preference will be given to the latest comprehensive annual renewable energy statistics that report both years on a consistent basis. The comparison will use the greatest available common precision rather than independently rounding both figures to whole gigawatts. The outcome depends exclusively on the reported annual additions, not whether China's solar industry achieves particular policy targets, whether total installed solar capacity increases, or whether solar electricity generation grows in 2027.