政策转向：光伏出口退税的终结与深层原因
2026年1月12日，中国财政部与国家税务总局联合发布公告，宣布自当年4月起，取消光伏产品的出口退税。这一决定标志着曾是行业重要支持政策的出口退税待遇的终结。从产业视角看，此举源于中国光伏产业已足够强大，不再需要国家层面的保护。而从财政角度审视，相当一部分光伏产业投资被地方政府为追求政绩而盲目推动，导致了严重的产能过剩和恶性竞争。因此，取消出口退税成为必然。
出口退税政策，简单来说，是一种国家层面的激励措施，旨在鼓励企业出口商品。在1980年代以前，中国的外汇收入主要由中央政府集中管理，大部分商品进口也需中央部委审批。1984年外贸体制改革后，地方政府被允许保留部分外汇收入，这促使它们积极参与进口。起初，地方政府虽也进口生产设备，但大量外汇被用于进口消费品，海南岛曾是典型案例，一年进口近8万辆汽车和300万台电视机，而当时中国外汇储备仅80多亿美元。这种政策漏洞导致了利用进口汽车进行倒卖和走私的现象持续存在。从1983年至1987年，地方政府动用160亿美元外汇进口汽车，而1988年全国外汇出口总额仅20亿美元，外汇储备急剧下降，国家面临破产风险。
Original English
hello, everyone, friday, january 28th, 2026. welcome to episode ten twelve of tax news. on january 12th, the ministry of finance and the state taxation administration jointly issued an announcement stating that starting from april this year, photovoltaic exports will no longer enjoy value-added tax rebates. why has the photovoltaic industry been stripped of this preferential treatment? from an industrial perspective, it's because the Chinese photovoltaic industry is strong enough and no longer needs protection. from a financial perspective, a considerable portion of photovoltaic investment is driven by local governments forcibly initiating projects for political gain, leading to excessive competition. therefore, the export tax rebate treatment has been revoked. let me first explain the export tax rebate policy. simply put, it's a national level policy that encourages companies to export goods. before the 1980s, foreign exchange earned by local governments and state-owned enterprises was centrally controlled by the central government. most commodity imports also required approval at the central ministry level. in nineteen eighty-four, China reformed its foreign trade system, allowing local governments to retain a portion of foreign exchange for their own use after obtaining import quotas. although they did import some production equipment, a large amount of foreign exchange was spent on consumer goods. taking Hainan island as an example, in one year, it imported nearly eighty thousand cars and three million television sets, which were then resold in the mainland. at that time, China's total foreign exchange reserves were just over eight billion US dollars. Hainan, which was only an administrative region, spent five hundred million. please read a news report for me: among the ninety-four units under the jurisdiction of Hainan, eighty-eight owned cars, and even schools and kindergartens were participating. in addition, Hainan also imported a large number of televisions, video recorders, and motorcycles. in nineteen ninety-five, the total number of cars imported by the country was equal to the total number of cars imported from nineteen fifty to nineteen seventy. after the Hainan incident, the phenomenon of using policy loopholes to resell and smuggle imported cars continued to exist. from nineteen eighty-three to nineteen eighty-seven, these local governments used foreign exchange to import a large number of cars, amounting to sixteen billion US dollars, equivalent to two crude oil corporations. a year later, China's total foreign exchange exports were only two billion US dollars, a decrease of more than six billion US dollars compared to the previous year. financial reserves were critically low, and the country was at risk of bankruptcy. in order to bring back export earnings, the order decided to enrich exports. at that time, among China's concentrated high-tech products in the military industry, only iron and steel were willing to pay. other industrial products or light industrial products had high costs, so they had to find a way to obtain price competitiveness in foreign exchange. therefore, starting in nineteen ninety-five, the central government introduced an export tax rebate policy to reduce enterprise production costs. who would pay for the tax rebates? this was the most crucial part of the policy. for the first three years, it was calculated according to the enterprise profit relationship. for example, if a central enterprise exported, the central government would bear the cost. if a local enterprise exported, the local government would bear the cost. but this brought another problem: although local governments could retain a portion of foreign exchange, the majority still had to be handed over to the central government, meaning that the more export performance, the more money the local government would lose. therefore, starting in nineteen eighty-eight, the country implemented some reforms in foreign trade. the central government bore most of the tax rebate costs. local governments also had to submit foreign trade plans annually to determine how much foreign exchange they would contribute. local governments could retain a portion of income exceeding the plan. from then on, local governments organized exports using a contract system. local governments acted as contractors for foreign trade until nineteen ninety-four when the foreign trade law was enacted. however, customs have remained under unified central management, belonging to the central government's authority for decades. afterward, the cost of export tax rebates has also been basically borne by the central government. in twenty sixteen, a reform nationwide to replace business tax with value-added tax meant that the VAT received by local governments is shared with the central government. as mentioned, the tax information section number one hundred one, one of the purposes of the central government adopting VAT was to transfer revenue from lower-tier cities to smaller cities. therefore, in terms of VAT export tax rebates, while maintaining transfer payments to local governments, this effectively means that the central government bears the cost of commodity exports, reducing the burden on local governments.出口退税的演变与地方政府的激励扭曲
从积极的视角看，出口退税政策能够降低企业生产成本，帮助国内企业保持价格竞争力并拓展国际市场。然而，另一方面，地方政府并未承担全部退税成本，深知存在套利空间。由于中央政府承担了退税费用，无论企业利润多少，出口至少能为企业贡献GDP。即使零利润甚至亏损销售，只要退税能弥补差额，企业仍能获利。因此，地方政府与企业形成了一种默契，共同利用中央政府的资金。过去十多年，许多地方政府和企业都在围绕“依赖中央政府资金”的思路发展。
光伏产业只是一个例子。中国光伏产业的早期发展，曾被形容为“三头在外”，即生产光伏板所需的高纯度多晶硅需从美国、德国进口，生产设备需从欧洲引进，且不到10%的光伏组件产品在国内消费，其余主要销往欧洲以消化过剩产能。2009年，中国启动“金太阳”工程，通过政府补贴发展国内光伏发电。据《北京商报》报道，该项目门槛极低，只要能获得并网屋顶许可即可申请，成功后可获得大量补贴。部分企业采购不合格材料，且项目完成后无人监管发电量，因此不惧被调查。结果是，整体补贴并未得到有效利用。财政部随后要求追缴“金太阳”项目2009-2011年近80%装机容量的补贴资金，可能超过70亿元。
Original English
from a positive perspective, the tax rebate policy can reduce enterprise production costs, helping domestic enterprises maintain price competitiveness and capture international markets. however, on the other hand, local governments did not bear the full cost of tax refunds, knowing the potential for arbitrage. since the central government was paying, regardless of how much profit companies could make, exports could at least contribute to GDP for companies, even with zero profit or even selling at a loss. as long as the tax refunds could cover the difference, they could still make some money. therefore, local governments and companies developed a tacit understanding to collectively exploit the central government funds. for the past decade or so, many local governments and companies have been developing strategies based on the idea of relying on central government funds. the photovoltaic industry is just one example. let's briefly review the development process of China's photovoltaic industry. many viewers may remember that the internal media once described China's photovoltaic industry as "three heads outside," meaning that the high-purity polysilicon used in photovoltaic panels had to be imported from the United States and Germany, the production equipment had to be imported from Europe, and less than ten percent of the photovoltaic panel products were consumed domestically, with the rest mainly sold to Europe to digest a surplus. in two thousand nine, China launched the "Golden Sun" project, developing domestic photovoltaic power generation through government subsidies. as for the effectiveness of this round of subsidies, please allow me to read a report from the Beijing Business Daily: a company told the Beijing Business Daily reporter that the threshold for the "Golden Sun" project was very low; as long as they could obtain grid-connected rooftop permits, they could apply, and if successful, they could receive a large amount of subsidy funds. some companies purchased substandard materials, and since no one supervised the power generation after the project was completed, they were not afraid of being investigated. as a result, the overall subsidies were not fully utilized. publicly, that shows that the Ministry of Finance demanded the recovery of subsidy funds for nearly eighty percent of the installed capacity of the "Golden Sun" projects from two thousand nine to two thousand eleven, possibly exceeding seven billion yuan.光伏产业的畸形扩张与国际制裁
国内产能过剩和海外市场出现问题时，中国光伏产品的售价仅为欧洲同行的生产成本的一半。此时，包括尚德电力（Suntech Power）和英利（Yingli Solar）在内的大多数公司都遭受了巨额亏损，仅靠出口退税得以生存。2012年起，欧盟对中国光伏产业发起反倾销调查，并最终决定征收64.9%的反倾销税和11.5%的反补贴税。随后，无锡尚德（Wuxi Suntech）这家中国最大的光伏制造商直接宣布破产。为支持国内光伏产业，国务院推出了新一轮补贴政策，这次补贴更为科学，实行“先发电，后付费”的模式。企业不能再依赖退税或虚报产量来获取补贴，而是必须老老实实地专注于技术研发以降低发电成本。至此，中国光伏产业的产业链和市场才真正实现了国内自主可控，并逐渐主导了全球市场。
中国掌握了光伏技术，同时也降低了国内产业发展的门槛。无论企业之前是养猪、种地还是种菜，只要有足够的现金流，就可以在国内购买整套生产线。例如，常州华塑（Changzhou Huasu）等企业帮助客户建设了超过三千条技术门槛低、有政府补贴的光伏自动化生产线。在中国光伏产业的推动下，到2022年，中国拥有600家自动化光伏企业，产业规模在2014年达到顶峰。当年，中国光伏新增装机容量占全球总量的50%，多晶硅产量占全球市场近80%。然而，缺乏技术壁垒的产业必然面临激烈竞争。与2012年相比，光伏发电成本下降了90%，但即使加上补贴，企业仍难以盈利。在某次光伏发电项目招标中，中标价格降至每瓦6.8美分，仅能勉强覆盖成本，几乎没有利润空间。
Original English
domestic production was discovered, and the overseas market entered problems. at this time, the selling price of Chinese photovoltaic products was only half of the production cost of their European counterparts. most companies, including Suntopower and N-type solar, suffered huge losses, and only survived by relying on export tax rebates. starting in twenty twelve, the EU launched an antidumping investigation against the Chinese photovoltaic industry and ultimately decided to impose a sixty-four point nine percent antidumping duty and an eleven point five percent antisubsidy duty. the following year, Wuxi Suntech, then China's largest photovoltaic manufacturer, directly declared bankruptcy. to support the domestic photovoltaic industry, the State Council launched a new round of subsidy policies. this time, the subsidies were more scientific, becoming a "generate electricity first, then get paid" system. companies can no longer rely on tax rebates or faking product claims to receive subsidies. they must honestly focus on developing technology to reduce power generation costs. before this, China's electricity industry and market were not aligned with resources. gradually, it transformed into a situation where all three were domestically sourced. only at this point did China truly possess its own photovoltaic industry chain, dominating the global market. China mastered photovoltaic technology while also lowering the entry barrier for domestic industrial development. companies, regardless of their previous business, e.g., pig farming or vegetable growing, could purchase complete production lines domestically as long as they had sufficient cash flow. for example, Changzhou Jintan Huaxing Plastics Co., Ltd. helped build over three thousand automated photovoltaic production lines with low technological barriers and government subsidies available. Chinese photovoltaic industry quickly boomed again. by twenty twenty-two, China had six hundred automated photovoltaic enterprises, and the industry scale peaked in twenty fourteen. in that year, China's installed photovoltaic capacity accounted for fifty percent of the global total, and its polysilicon industry accounted for nearly eighty percent of the global market. China's newly added installed photovoltaic capacity reached two hundred seventy-seven gigawatts, exceeding the sum of the rest of the world. industries without technological barriers inevitably face fierce competition. compared to twenty twelve, photovoltaic power generation costs decreased by ninety percent, but even with subsidies, companies still struggle to make a profit. after a photovoltaic power generation project tender in China, the winning bid dropped to sixty-eight cents per watt, barely covering cost and leaving almost no profit. however, simultaneously, for the sake of political achievements, local governments continue to invest eight hundred billion yuan to build more photovoltaic factories. this domestic market competition in China's photovoltaic industry also impacted the international market, effectively crushing foreign competitors. from twenty twenty-three onwards, European experience severe overcapacity and photovoltaic production, with virtually all photovoltaic companies announcing layoffs or factory closures. the global solar energy industry has almost entirely become dominated by Chinese companies. in twenty twenty-four, China exported two hundred thirty-six gigawatts of solar components, a thirteen percent increase compared to twenty twenty-three. domestic installations remained... [unintelligible Chinese phrase].地方融资平台：政绩驱动下的债务扩张
与此同时，为了追求政绩，地方政府持续投入巨资建设光伏工厂。国内市场竞争也影响了国际市场，从2023年起，欧洲经历了严重的产能过剩和光伏产品滞销，几乎所有光伏公司都宣布了裁员或工厂关闭。全球太阳能产业几乎完全被中国企业主导。2024年，中国出口了236吉瓦的太阳能组件，同比增长13%。国内新增装机量依然可观。2025年3月，中国多个省份的太阳能出口贸易额超过1亿美元，11个省份超过1000万美元。虽然市场份额在增加，但企业盈利却越来越少。2024年，太阳能产品总出口额增长33%，但总交易额却下降了33%，这表明中国太阳能产品价格大幅下跌，以消化过剩产能。
各组件数据的跌幅更为惊人。例如，太阳能电池出口量增长了10%，但总交易额却下降了30%。若从2020年计算，中国出口太阳能组件的价格跌幅已超过60%。许多观众可能还记得，曾有新闻报道称，因为中国太阳能板价格比普通建材还便宜，许多欧洲人直接用太阳能板修补房屋。这些都是由中国中央政府通过税收补贴的产品。换言之，我们都在为这种内部竞争和无人盈利的价格体系买单。2023年，中国太阳能产业总产值接近1.4万亿元人民币，但到2024年，却降至7700亿元人民币。几乎所有上市太阳能公司的营收都大幅下滑，整个行业账面亏损244亿元人民币，潜在亏损按正常市场原则衡量是不可估量的。
如果一个产业无法盈利，内部的“内卷”将从小范围开始，缺乏竞争力的公司将持续破产，最终只剩下少数高水平的领先企业重新盈利。2012年，一家中国领先的太阳能公司遭遇市场挑战并破产，这表明当时市场原则仍然有效。然而，如今这一原则已失效。如前所述，中国太阳能公司数量从2022年的60多万家，预计到2025年将达到124万家，几乎翻倍。太阳能公司在逆势扩张，债务也在不断增加。截至2023年底，中国117家A股上市的太阳能公司总负债接近1.99万亿元人民币，其中37家公司负债超过100亿元，11家超过500亿元，3家超过1000亿元。截至2024年3月，总负债已增至2.32万亿元人民币，比上一年增加超过3400亿元人民币。若包含未上市公司，数字将更高。太阳能公司之所以能在亏损的情况下大量借款，得益于地方政府的支持。地方政府认为，让这些公司留在当地能创造GDP，而太阳能产业成熟的技术能快速提升产值，从而在其任期内创造政绩。至于补贴成本，则完全依赖中央财政资金，为地方政府提供了看似无限的资金。即使市场不景气，太阳能公司也不会主动退出，而是希望其他地方政府先失败，自己能坚持到最后。这与过去借钱搞“面子工程”的心态相似。地方政府并非不相信自己能赚钱，而是认为其他地方政府的融资平台会先出问题。
Original English
however, simultaneously, for the sake of political achievements, local governments continue to invest eight hundred billion yuan to build more photovoltaic factories. this domestic market competition in China's photovoltaic industry also impacted the international market, effectively crushing foreign competitors. from twenty twenty-three onwards, European experience severe overcapacity and photovoltaic production, with virtually all photovoltaic companies announcing layoffs or factory closures. the global solar energy industry has almost entirely become dominated by Chinese companies. in twenty twenty-four, China exported two hundred thirty-six gigawatts of solar components, a thirteen percent increase compared to twenty twenty-three. domestic installations remained... [unintelligible Chinese phrase]. in march twenty twenty-five, Chinese provinces had solar export trade volumes exceeding one hundred million yuan, and eleven provinces exceeded ten million yuan. while market share is increasing, companies are earning less and less money. although the total export volume of solar products increased by thirty-three percent, the total transaction value decreased by thirty-three percent. this indicates a significant price drop in Chinese solar products to absorb excess capacity. the data for various components is even more astonishing. for example, the export volume of solar cells increased by ten percent, while the total transaction value decreased by thirty percent. if calculated from twenty twenty, the price of China's export solar components dropped by over sixty percent. many viewers may still remember a ridiculous news story: because the price of Chinese solar panels was cheaper than ordinary building materials, many Europeans directly used solar panels to repair their homes. these are products subsidized by the Chinese central government through tax rebates. or rather, we are all paying taxes to subsidize this internal competition and the prices that no one makes money. in twenty twenty-three, the value of China's solar industry was nearly one point four trillion yuan, but in twenty twenty-four, it became seven hundred and seventy billion yuan. the revenue of almost all solar listed companies has fallen sharply, and the entire industry has a book loss of twenty-four point four billion yuan with potential losses that are measurable according to normal market principles. if an industry cannot make money, internal competition will begin on a small scale, and uncompetitive companies will continuously go bankrupt. and finally, a few high-level leading companies will remain and start making money again. in twenty twelve, a leading Chinese solar company encountered market challenges and went bankrupt the following year, indicating that market principles were still effective at that time. however, this principle is now ineffective. as mentioned earlier, the number of Chinese solar companies, which was six hundred thousand in twenty twenty-two, is finally expected to reach one point two four million in twenty twenty-five, nearly doubling. solar companies are expanding against the trend, and their debt is also increasing. by the end of twenty twenty-three, the total debt of one hundred seventeen A-share listed solar power companies in China reached nearly one point nine nine trillion yuan, with thirty-seven companies having debt exceeding ten billion yuan, eleven exceeding fifty billion yuan, and three exceeding one hundred billion yuan. by the end of march twenty twenty-four, the total debt had increased to two point three two trillion yuan, an increase of over three hundred forty billion yuan compared to the previous year. if unlisted companies are included, the number would be even higher. solar power companies are able to borrow huge money while losing money thanks to the support of local governments. local governments believe that keeping these companies in their area will create GDP, and the mature technology of the solar power industry can quickly increase production value, thus creating political achievements during their term of office. as for the subsidy costs, they completely rely on the central government's treasury funds, providing local governments with seemingly unlimited funds. even with the market downturn, making solar power companies will not voluntarily withdraw, hoping other local governments will fail first and they can hold out until the end. this is similar to the mentality of borrowing money to build vanity projects in the past. local governments don't believe they can make money, but rather that other local government financing platforms will encounter problems first. however, after twenty twenty-four, the pressure became too great, and some local governments finally realized the seriousness of the situation. they were convinced the final outcome would not be that other local government investments would fail first, but that the entire industry would collapse. at this point, the government became anxious to recover the subsidy funds.重塑市场机制与终身问责
然而，到了2024年后，压力变得过大，一些地方政府终于意识到情况的严重性。它们确信，最终结果并非是其他地方政府的投资会先失败，而是整个行业都会崩溃。此时，政府开始急于追回补贴资金。例如，2026年1月23日，焦作市人民政府发布行政决定，责令ST美邦（ST Meibang）向武陟县返还5.1亿元财政补贴和建设资金，并支付5100万元违约金。广西南宁市也宣布，因某公司未能按时完成回购，终止了某光伏项目投资协议，追回了1.4亿元投资款。2026年1月5日，一家苏州的国企子公司因其投资的旅游公司未能按时完成回购，向其索赔3.26亿元。
过去几年，《税务新闻》工作室一直关注地方债务问题，广泛报道了许多文旅项目和无效基建项目。一些观众可能已经注意到，七八年前地方政府借钱发展旅游业的逻辑，与近几年投资“政绩工程”的逻辑如出一辙。这并不奇怪，因为投资文旅项目和投资政府“政绩工程”的公司，很可能属于同一批投资者。据报道，自2024年1月起，超过200家国有企业转型为产业投资公司，2024年全国新增的国有企业也多涉及产业投资。许多国有资产被转移，最终落入产业投资平台手中。例如，去年4月，山东某国企无条件将旗下旅游公司股权转让给产业投资平台。再以山东临沂为例，这是一个典型的国有企业债务公开化的问题城市。
2025年3月24日，《大众新闻》报道，一位来自内蒙古的张先生于2023年2月11日购买了山东临沂国信基础设施建设有限公司发行的债券。该债券资金用于建设莒南县妇幼保健服务中心，并由莒南县城建投资集团有限公司担保，这是一种典型的多重担保，极具误导性。张先生表示，按协议两年应全额还本，但每月只还1%。他向莒南县政府反映后，工作人员称还款正在处理，每月只能还1%。张先生无奈表示，不要利息了，只求2025年初能还本。当时银行定期存款利率已普遍跌破1.5%，而老赖集团却以9%的年利率借款，甚至无力还本。这就是该集团金融平台的最终困境。
《科教晚报》2026年的报道《改革中探索产业赋能发展——O.ONE集团的转型之路》指出，在成都，O.ONE集团已从政府投资融资平台，转变为产业运营平台。这种创新模式盘活了资本平台和区域资源，将优质项目引入三大产业园区，使成都成为产业领导者的转型路径日益清晰。然而，在不同旗号下，这些集团的人依然将资金投入政府背书的产业，继续追求政绩。这种投资心态下，任何事情都可能发生。因此，国家必须切断过度的出口退税，打破地方融资平台的内部竞争机制。2024年12月，部分中国光伏产品的出口退税率已从13%降至9%，而现在完全取消退税是完全可以预期的。至于取消退税是否会影响中国光伏产业，我们可以保持乐观。光伏产业的发展曾受到欧盟反倾销制裁的严重影响，迫使中国企业停止依赖廉价劳动力和政府补贴来盈利。在领先企业破产后，中国光伏产业反而建立了技术优势。这一轮的重塑，中国企业应明白市场竞争的目的是盈利，而非依赖运气。
与旅游业相比，中国光伏产业具有更大的优势。它在国际市场上没有短期竞争对手。取消出口退税对其而言是好事。过去，地方政府利用投资平台投资水上乐园、豪华建筑等项目。如今，地方政府投资平台之间的“内卷”逻辑是相同的。初期，它们利用中国成熟的产业环境快速创造GDP和政绩。下一阶段，它们依赖中央政府有限的信用，希望中央财政能救助它们。要让市场机制重新发挥作用，唯一的方法是让政府投资平台的负责人对其决策终身负责，即使离任后也要对其任期内的决策负责。
关于您之前讨论的中国新能源汽车，以及您对政府投资和补贴的积极作用的认可，政府投资平台应承担终身问责。这是否会打击地方政府的积极性？政府投资是必要的，投资也带有一定风险。然而，由于地方官员从成功的投资中获益，即使在投资成功与否未知的情况下，他们已基于GDP和就业数据获得了政治成就。他们有义务为失败的投资承担部分责任。没有人要求决策者个人赔偿损失，但不能说失败的投资无人负责。至少，投资失败应被记录在他们的个人档案中。另一方面，实施终身问责制并不一定意味着对负面结果负责，它实际上可能提供更多试错空间。例如，某项投资初期未赚钱，但后期开始盈利，仍可视为相关官员的成就。如果某些官员认为终身问责制对其不利，这表明他们清楚自己任期内的所作所为。
市场经济的最大好处并非必然是选出最好的公司，因为随机性始终存在，许多优秀的公司也可能在竞争中失败。然而，市场经济必须能够淘汰不具竞争力的落后公司，迫使其破产。目前，融资平台与过去的市政投资公司或现在的产业投资公司一样，首先建立地方政府的信用，然后间接利用中央政府的无限信用，拒绝进入市场。这是最大的问题。如果终身问责制能够抑制这些本不应存在的投资，那应该被视为好事。
最后，节目播出前，一位观众的提醒让我注意到，《聚焦24小时》2024年7月24日的采访包含对地方政策的长期批评。让我们直接看看主持人涉及的政治偏见：将可见的项目视为可见的政绩。即使财政能力紧张，他们仍然推动高强度的投资。这一事件也给我们留下了深刻的教训。我们坚持将安全和重大项目评估置于优先地位，将必要性、可行性和财务可持续性作为先决条件，以防止类似事件发生。但经过一段时间的整顿，这种因严重政治和官员失误以及盲目累积而造成的工业园区项目正在缓慢改善。然而，由此造成的资源浪费和财政负担将需要很长时间才能缓解。地方产业发展是可以理解的，但不能以此为借口进行盲目扩张，建设缺乏必要条件且不适合当地的工业园区。结果往往是由于脱离实际而导致建设和运营困难，给人民和企业造成损害。我们应该谨慎行事，尽力而为，坚持从实际出发，遵循规则，为人民争取政绩，并取得实实在在的成果。感谢收看。第十二期节目到此结束，我们下期再见。