「1987版黑色星期一」重演:抱团交易逆转、流动性冲击,之后发生了什么?

Odaily星球日报Published on 2024-08-06Last updated on 2024-08-06

Abstract

危险在于:暴跌可能自我强化,演变成信贷紧缩。

原文作者:赵颖,华尔街见闻

「 1987 版黑色星期一」在昨日再度上演,全球金融市场崩盘式暴跌,到处充斥着熔断、熊市、历史记录等字眼。

日经 225、东证指数双双暴跌逾 12% ,盘中多次触发熔断机制;台股创下 1967 年以来最大跌幅,韩国创 2008 年以来最大跌幅,道指跌超千点与标普齐创两年最大跌幅,富途、Fidelity 等纷纷警告交易故障。

上一次全球市场经历如此惨痛的洗礼,还是在 1987 年 10 月 19 日的股灾。

当时亚太股市纷纷跳水,日经指数下跌 14.9% ,恒生指数暴跌超 40% ,新西兰股指甚至一度暴跌 60% 。美国市场也陷入一片混乱,道指日内暴跌 22.6% ,标普 500 指数暴跌 30% ,全球股市约 1.71 万亿美元灰飞烟灭。

除了触目惊心程度相近之外,两次暴跌的导火索也类似,套利交易和程序化交易发生「大逆转」。以史为鉴,接下来会发生什么?美联储会再度出手「救市」吗?

「1987版黑色星期一」重演:抱团交易逆转、流动性冲击,之后发生了什么?

「 1987 版黑色星期一」

回顾 1987 年的美股走势, 10 月 14 日美国政府公布的贸易赤字比预期的要大,美元也随之贬值,市场开始走低。

10 月 16 日周五,美国众议院提出立法,消除与融资并购相关的某些税收优惠的报告,美股下跌加剧,并为接下来一周的动荡埋下了伏笔;

10 月 19 日周一开盘,人们恐慌地看到,市场中卖盘远大于买盘。由于差异过大,在最初的一小时内,众多场内做市商甚至不提供做市报价。

美国 SEC 后来指出,到 10: 00 ,标普 500 成份股仍有 95 支没有开盘;华尔街日报则指出, 30 支道琼斯成份股票中,有 11 支无法开盘成交。

与此同时,随着股指期货和股票之间出现较大套利空间,一批交易机构开展了套利交易,而随着股市继续跳水,大量套保盘进一步在股指期货市场上做空指数合约,这反过来继续推动股指的跳水。

收盘时,道琼斯指数暴跌跌幅 22.76% ,创造了 1929 年以来最大的跌幅数字。

10 月 20 日周二开盘前,美联储发表简短声明,并宣布「紧急降息 50 基点 + 量化宽松」救市:

美联储今天坚守其作为国家中央银行的职责,确认其愿意充当流动性来源,支持经济和金融体系。

市场也在美联储表态当天企稳,美股早盘继续下跌,芝加哥期权交易所和商品交易所中午暂停交易,午后恢复交易,随后有所反弹。

10 月 21 日,市场开始收复部分跌幅。

「1987版黑色星期一」重演:抱团交易逆转、流动性冲击,之后发生了什么?

套利、程序化交易平仓引爆

与 1987 年类似, 2024 年的「黑色星期一」也是由一场完美风暴触发。

彼时的美股自 1982 年以来一直处于大牛市,人们认为已经到了进行调整的时候了,而如今 AI 热潮助推的这波美股科技股走牛也让投资者「寒若禁蝉」。

其次是抱团交易逆转, 1987 年这场股灾中「程序交易」被认为是罪魁祸首之一,投资组合的交易程序抛售股票,从而产生了多米诺骨牌效应。

而最近的股市暴跌部分源于,美日利差收窄引发「套利交易」逆转。日本央行上周意外加息,上周美联储会后释放降息信号,联储 9 月降息几乎被完全定价,此前汇市最受欢迎的「卖日元,买美元」套利交易「魅力」不再,投资者开始将手中的美元资产换回日元。

与此同时, 1987 年崩盘前的那个星期五也出现了「三巫日」——股票期权、股指期货和股指期权合约同时到期,这导致了周五交易最后几个小时的严重不稳定,并将动荡延续到周一。

最后,分析将此次大幅下跌归咎于「群体性歇斯底里」,每次市场大跌时,投资者的从众心理会加剧下跌。

美联储会再度出手「救市」吗?

以史为鉴,美联储会采取什么行动?

应对 1987 年的市场崩盘,美国「紧急降息」、设立熔断机制,并提供流动性来救市。

为了减缓金融市场下跌并防止对实体经济产生溢出效应,美联储迅速采取行动向金融系统提供流动性,通过量化宽松政策向经济注入了数十亿美元资金。

「1987版黑色星期一」重演:抱团交易逆转、流动性冲击,之后发生了什么?

同时,时任美联储总统的格林斯潘宣布「紧急降息 50 个基点」,将联邦基金利率从周一的 7.5% 以上降至周二的 7% 左右。

「1987版黑色星期一」重演:抱团交易逆转、流动性冲击,之后发生了什么?

此外,监管机构还首次引入熔断机制,以防止因程序化交易而导致的市场崩溃。一旦出现股票市场异常下跌或上涨的情况,交易就会立即停止。

大跌将如何收场?

分析认为,最坏的结果可能是 2008 年的重演,但这种情况似乎不太可能发生。虽然去年一些美国大型银行因在政府债券上押注失误而倒闭,但银行的杠杆率比以前低得多,而且由于私人信贷承担了过去银行承担的大部分风险,银行体系受流动性危机的影响也较小。巨额亏损是有可能的,私人基金也可能会陷入困境,但这需要时间,而且不会引发同样的系统性危机。

理想的情况是,股市过度波动会像 1987 年那样逐渐平息,而不会造成更大范围的麻烦,预计这种平息过程比 1987 年更加缓慢。AI 狂热可能导致股价进一步下跌,即使从 6 月份的高点下跌了 30% ,英伟达的股价今年仍上涨了一倍。但市场已经更接近正常水平,纳斯达克 100 指数今年迄今仅上涨了 6% ,标普指数上涨不到 9% 。

「债券卫士」之父 Yardeni 认为:

市场大跌的危险在于,暴跌可能自我强化、演变成信贷紧缩。可以想象,这种套利交易平仓可能会演变成某种金融危机,进而导致衰退。

不过,他强调,他个人预测不会最终产生这种结果。

原文链接

Related Reads

Idle Macs Can Also Make Money? An Overview of Eigen Labs' Decentralized AI Inference Network Darkbloom

AI inference is becoming a crucial layer of internet infrastructure, yet it remains largely dependent on costly, capacity-limited centralized systems with potential security risks. Meanwhile, millions of powerful computers sit idle globally. Eigen Labs' Darkbloom network aims to utilize this idle capacity by enabling distributed AI inference on Mac computers, specifically those with Apple Silicon chips. Darkbloom's architecture consists of three components: users who send inference requests, a coordinator (operated by Eigen Labs) that routes these requests, and providers (Mac owners) whose machines run the models and return outputs without being able to see the request content. The system prioritizes privacy through a hardened provider process, software integrity checks, and hardware-supported attestation based on Apple's security architecture to ensure verifiable privacy. Economically, Darkbloom differs from traditional models. It leverages existing hardware, with marginal costs primarily driven by electricity, allowing it to offer pricing roughly 50% lower than major API aggregators. Providers keep 100% of the inference revenue, and the project does not rely on token subsidies; earnings come solely from real AI inference demand. However, early-stage earnings are modest, with top providers currently earning under $6 per day, influenced by factors like hardware specs, uptime, and network demand. The network currently supports models like Google's Gemma 4 and OpenAI's GPT-OSS via OpenRouter. To participate as a provider, users need an Apple Silicon Mac running macOS 14 or later, must install the Darkbloom provider software, and keep the machine online with a stable internet connection.

marsbit8m ago

Idle Macs Can Also Make Money? An Overview of Eigen Labs' Decentralized AI Inference Network Darkbloom

marsbit8m ago

Which Crypto Sectors Have Been "Eaten" by AI Agents?

The article examines which crypto sectors have been increasingly dominated by AI Agents and which remain human-centric. In certain high-speed, efficiency-driven areas, AI Agents have taken clear control. This includes derivatives/perpetuals trading, where bots outperform humans significantly (e.g., a contest showed 0% of AI Agents were liquidated vs. 43% of humans), arbitrage/MEV extraction, and yield optimization (with ~68% of new DeFi protocols in Q1 2026 featuring autonomous AI Agents). Spot trading and portfolio optimization are also seeing heavy Agent adoption. However, the shift is not universal. In "battleground" sectors, both Agents and humans coexist. In prediction markets, Agents dominate short-term arbitrage, but humans still outperform in long-term, nuanced judgment calls. In DeFi lending, while liquidation is automated, core deposit/borrow decisions remain largely human-driven. Sectors still firmly led by human activity include stablecoin payments and card-based spending (driven by real-world economic activity and remittances) and wallets, which serve as the crucial human-verification and approval layer. The rise of Agents increases the need for robust human-Agent verification layers. Projects like World/AgentKit, t54, Self Protocol, and Kite AI are building infrastructure to create trust, security, and accountability by binding Agents to verified human identities. In conclusion, while AI Agents have decisively "eaten" speed and optimization-focused crypto sectors, human judgment, trust, and real-world context remain dominant in areas that create broad economic value, such as payments and identity. The future likely involves a symbiotic relationship where Agents require human verification and oversight to operate effectively.

Foresight News14m ago

Which Crypto Sectors Have Been "Eaten" by AI Agents?

Foresight News14m ago

After Rising 11 Times in a Year, Micron's Earnings Report Becomes a Stress Test for the AI Memory Market

**Micron's Upcoming Earnings: A Crucial Test for the AI Memory Rally** Investors in AI memory stocks face a critical moment on June 24th, when Micron Technology reports quarterly earnings. The stock, having surged approximately 11-fold from $103 to $1,134 over the past year, carries immense market expectations. Wall Street consensus forecasts a staggering ~932% year-over-year jump in EPS to around $19.72 and ~270% revenue growth to ~$345 billion, largely driven by sold-out HBM (High Bandwidth Memory) capacity through 2026. Analysts have aggressively revised estimates upward over the last 90 days, with EPS expectations rising 68%. This creates a high bar: even strong results risk a sell-off if they fail to meet these elevated projections. Notably, price forecasts from institutions like Citi (predicting ~200% DRAM price increases in 2026) are already among the most bullish on Wall Street, not conservative. The key metric to watch is gross margin, guided to a record ~81%. Such peak profitability raises questions about sustainability in the historically cyclical memory sector. While management has signaled continued strength, the stock's direction post-earnings will likely hinge more on forward guidance for the next quarter and details on HBM capacity expansion for 2027, rather than the already-anticipated stellar past results. The report represents a major pressure test for the high-flying AI memory trade.

marsbit18m ago

After Rising 11 Times in a Year, Micron's Earnings Report Becomes a Stress Test for the AI Memory Market

marsbit18m ago

Trading

Spot
Futures
活动图片