摩根大通减持亚马逊 56.2%:科技股机构持仓异动与市场情绪分析
📊 本文涉及标的速览
摩根大通(JPMORGAN CHASE & CO)于美东时间 2025 年 12 月 31 日披露的 13F 持仓报告显示,其对亚马逊(NASDAQ: AMZN)的持仓环比大幅削减 56.2%,持股比例从上一报告期约 3.4% 降至 1.49%。这一减持规模在大型机构对蓝筹科技股的仓位调整中极为罕见。与此同时,同期的挪威主权财富基金 Norges Bank 将 AMZN 持仓提升 13.9% 至 1.32%,资管巨头 T. Rowe Price Associates 亦减持 8.4% 至 1.22%。三方机构仓位方向分歧,市场解读出现明显分化。
机构持仓异动:谁是对手盘
13F 报告每季度披露一次,是观察机构资金结构性流动的核心窗口。本次摩根大通对 AMZN 的减持并非孤立事件——同季持仓变动中,该行对英伟达(NASDAQ: NVDA)的减持幅度同样高达 53.3%,持股比例从 3.7% 降至约 1.88%,T. Rowe Price 也同步减持 NVDA 5.7% 至 1.54%。
这一 패턴 引发市场对"机构轮动"逻辑的讨论:摩根大通在科技板块内部进行了系统性减持?还是在高估值压力下执行止盈策略?从持仓比例的绝对值观察,摩根大通的 AMZN 持仓仍维持在 1.49%,这一比例仍属于该公司前二十大股东序列,持仓并非清仓式减持,性质上更接近"边际减仓"而非"看空离场"。
挪威主权基金 Norges Bank 的反向增持则提供了一种解读维度。作为长期耐心资本,Norges Bank 在本周期选择加仓 AMZN,通常代表其对亚马逊长期基本面的认可。近年来 Norges Bank 在科技板块的布局持续扩大,其持仓变化往往被视为全球主权基金对科技资产配置的晴雨表。
T. Rowe Price 的减持幅度相对温和,仅为 8.4%,属于正常的组合再平衡行为。大型主动型基金管理人基于业绩基准约束进行仓位微调,在季度间较为常见,不宜过度解读为看空信号。
市场反应:AMZN 承压,NVDA 反弹
持仓报告披露当日,AMZN 股价收跌 1.08% 至 $267.22,为当日快照中主要科技股里唯一收跌的标的。同日英伟达(NVDA)虽也遭受摩根大通大规模减持,但股价反而大涨 4.39% 至 $235.74,两者走势形成显著分化。
这种分化背后存在多条逻辑线索。英伟达当日的大涨与 AI 算力需求持续高企的市场叙事密切相关,基本面预期的强劲对冲了机构减持的短期压力。而亚马逊则面临多重定价压力:电商业务利润率回归常态、云服务 AWS 增速市场预期降温、以及零售业务竞争格局加剧,均对估值形成压制。机构减持的负面信号在基本面边际承压的环境下更容易被放大。
从资金面角度,$267.22 的收盘价意味着 AMZN 在 2025 年以来的涨幅部分回吐。比特币(BTC)同日上涨 2.46% 至 $81,281,狗狗币(DOGE)上涨 2.73%,加密资产整体走强形成风险偏好扩张的对比,更凸显 AMZN 当日走势的疲弱。
历史对照:大型机构大幅减持后股价表现
回顾近五年大型机构对头部科技股进行超过 50% 减持股的案例,可以提供一定的参照框架。
2023 年第一季度,挪威主权基金曾单季减持苹果(AAPL)约 12%,其后苹果股价在季度内基本持平,未出现持续性下跌,关键在于其他机构资金承接了抛压。2022 年第四季度,多家对冲基金减持 Meta Platforms 超过 40%,其背景是元宇宙业务亏损扩大与数字广告宏观压力共振,Meta 股价在随后六个月内累计下跌逾 30%,彼时的减持具有前瞻性基本面的意义。
回到本次 AMZN 案例,摩根大通的减持发生在 2025 年下半年科技股估值分化的背景下。与此前案例不同的是,AMZN 的核心业务指标——AWS 收入增长、订阅服务增长——在公开披露中仍保持正向趋势,这降低了"前瞻性看空"的可能性。更可能的解释是摩根大通基于组合管理需求,在 AMZN 持仓已超基准权重的情况下进行边际减持。
T. Rowe Price 的减持幅度 8.4% 远低于摩根大通的 56.2%,说明不同管理人间对 AMZN 的定价存在显著分歧。Norges Bank 的逆势加仓进一步印证了这一分歧,并非单边看空格局。
关键观察点与风险因素
对于关注这一持仓异动的投资者,以下几个维度值得持续跟踪。
摩根大通的后续动作:13F 报告披露的持仓数据存在约 45 天的滞后性,2025 年 12 月 31 日的数据实际反映的是 2024 年 10-12 月间的交易。下一份持仓报告(截至 2026 年 3 月 31 日,预计于 2026 年 5 月中旬披露)将是判断减持是否持续的关键依据。若摩根大通在 2026 年第一季度进一步削减 AMZN 持仓,将加强"机构系统性降低科技敞口"这一判断的可靠性。
Norges Bank 加仓的持续性:作为全球最大的主权财富基金之一,Norges Bank 的季度持仓变化向来被视为长期配置方向的领先指标。若该基金在连续两个季度内加仓 AMZN,将对估值底部形成一定支撑。
AWS 业务增速与盈利质量:亚马逊的估值核心逻辑正从"电商规模增长"向"AWS 与 AI 服务货币化"迁移。2025 年下半年以来,AI 基础设施投资持续高企,但竞争格局——微软 Azure、谷歌 GCP 的追赶——对 AWS 市场份额形成压力。投资者需关注 AWS 收入增速是否出现边际下滑,这将直接影响市场对 AMZN 的估值重定价。
市场整体风险偏好:当日比特币突破 $81,000、标普 500(SPY)收涨 0.62% 至 $748.17,整体风险资产环境偏暖。AMZN 的逆势下跌在这种背景下更具警示意义——在市场整体向好时出现单独承压,往往意味着该标的正面临超出系统性的定价压力。
综合来看,摩根大通对 AMZN 减持 56.2% 是一次显著的机构仓位调整,但考虑到 Norges Bank 的反向加仓以及 T. Rowe Price 的有限减持,断言"机构看空亚马逊"仍为时过早。这一异动更准确地反映了不同机构对科技股估值分化的应对策略分化——部分资金在高估值环境下执行止盈,另一部分长期资金则在回调中逆势布局。后续 13F 数据与财报季AWS增速将是验证这一判断的核心窗口。
🗳️ 你怎么看?
读完本文后,你对所涉标的的短期走势倾向:
0 人已投票 · 同一 IP 仅可投一次
💬 社交媒体讨论
来自 Hacker News、StockTwits 等公开来源的真实评论摘录。展示给读者补充全网视角,不代表本站观点。
Funnily enough it is actually illegal in Europe and so our laid off German employees were reinstated by court order, several other countries as well. Amazon presumably ran into the same problem, or was smart enough to know they would: they give plenty of notice to European employees and follow local laws. https://www.reddit.com/r/cscareerquestionsEU/comments/1qpni5...
The Silicon Furnace Is Running Too Hot Not To Touch $NVDA $AMZN $META $AVGO https://talkmarkets.com/article/the-silicon-furnace-is-running-too-hot-not-to-touch-1778811615
I've personally had people talk about token leaderboards at their work. Amazon and Meta did have ones, but I'd take with a decent grain of salt. We all know it's such an insanely gameable metric you'd be insane to actually use it...
It's trivial to find reviews and benchmarks of DS4 online. Also, there are benchmarks in the article. Here's one of the top hits: https://forums.developer.nvidia.com/t/fully-custom-cuda-nati... Bizarre comment; sounds like "How do you know Porsches are fast? Did you drive one?"
I'd put one of these in my backyard for free access[0] > The boxes contain 16 Nvidia GPUs, 4 AMD CPUs, 4 terabytes of memory, and a cooling system. [0] https://www.fastcompany.com/91539193/home-side-mini-data-cen...
These problems feel like they should be solveable. Communities should set strict noise limits and fine the heck out of data centers for exceeding them. Data centers should be required to provide their own power, like this datacenter [1] in Ireland (and ideally a large portion of it should be renewable). They should also be required to minimize water usage, like this Microsoft d…
$TSLA $NVDA $GOOGL $MSFT $ORCL One of the most dystopian pics you will ever see. Idiot humans cheering and clapping for the technology that will replace them.
Better software means nearly nothing at the end of the day. The software that gets used is the metric that really matters. You can write 'perfect' software that runs at 100% efficient and never makes a mistake, but if no one ever downloads it and uses it, you've just engaged in a bout of intellectual masturbation. And honestly I've seen the 'write better software' people compla…
Older models are locked to Tesla vehicles. Tesla has regional restrictions in many parts of the world. You also never know when there could be another update and your region becomes one of those that has these restrictions.
$SPY $NVDA $BA $TSLA $QQQ Why do they gotta be so vague?Where's the details?
There are some restrictions. I had the foolish idea of installing a Tesla charger at home to charge my Bolt. I’ve been unable to ever use it. The wall charger works fine with Teslas. My car and adapter charge fine at Tesla superchargers. But the home Tesla charger refuses to charge my Bolt. (Yes I disabled vehicle restrictions and tried all sorts of combinations of settings for…
Great to find this narrow focused thing: > We support the following backends: Metal is our primary target. Starting from MacBooks with 96GB of RAM. NVIDIA CUDA with special care for the DGX Spark. AMD ROCm is only supported in the rocm branch. It is kept separate from main since I (antirez) don't have direct hardware access, so the community rebases the branch as needed. > This…
📊 想看实时数据?
本文涉及标的的实时价格、舆情指数均可在监控面板查看。 设置告警规则后,命中即时通过 Telegram / 邮件推送。
💬 读者评论