当前的市场动荡是否正在呼应1997和2013？
现在正值开学季，我们回到办公桌前，却发现离开时未解决的紧张局势依然存在。没有崭新的铅笔盒，取而代之的是地缘政治和宏观经济风险持续上升，推高全球债券收益率，而股市仍接近历史高位，科技股掩盖了其他领域的疲软。同样的根本问题依旧存在：这一切何时会崩溃？
虽然不幸的是我们没有水晶球，但回顾历史或许会有所帮助。我们专有的宏观时机模型MacroScope通过审视过去的市场机制，寻找有助于解读当前状况的模式，显示当前环境与1997五月最为相似。那时距离泰国货币贬值并引发亚洲金融危机还有五周。
2013,二月（美联储缩减恐慌前三个月）位居第二。虽然今天的条件并非与那些时期完全吻合，但与2026九月共享的一点是，当时普遍存在的风险已广为人知，但尚未造成破坏，使得投资者大多能够忽视这些风险。
亚洲的新兴压力最初被视为区域性且可控的问题。美国经济当时正享受“金发女孩”时期，低失业率、通胀受控和生产率提高罕见地结合在一起。直到十月底，道琼斯指数暴跌7.18%，熔断机制首次暂停交易，危机才波及华尔街。
标普500指数在1997结束时仍上涨31%，但最初的动荡引发了一场波及全球的系统性冲击，并持续了整个1998年。[1]
2013环境从另一个角度说明了类似观点。每个人都知道美联储的印钞机最终会关停，但市场依然攀升。央行多年来一直购买债券，以压低借贷成本，并推动投资者将资金从安全资产转向风险资产。增长疲软，支出削减即将带来影响，因此当清算在五月到来时，无人应真的感到意外。
那么，这段历史教训对今天有多大参考价值？我们当前的立场倾向于谨慎悲观，而非看空。无论是投入到人工智能的万亿资金、伊朗战争，还是美国公共财政状况，这些明显的候选因素中，没有哪一个看起来可能单独触发迫在眉睫的崩溃。
财政受限的美国ZF、我们认为即将加息的美联储，以及仍以相当良性结果定价的估值，所有这些因素共同使得市场几乎没有犯错余地。这些因素相互交织并普遍相互加强，表明不需要太大的冲击就能暴露潜在的脆弱性。
或许令人稍感安慰的是，模型指出2011,八月（美国失去AAA信用评级、欧元区债务危机威胁单一货币集团）是与当前环境最不相似的时期。
截至撰写本文时，华尔街的恐慌指标VIX指数接近其年内低点14.2，于8月14日触及今年最低水平。10月到期的VIX期货已小幅升至10月的19和11月的19.7，预期美国中期选举前的波动性将上升。然而，这距离反映市场严重压力的心理阈值30仍有相当距离。
7月9日，投资者对指数预期波动率与对成分股预期波动率之间的差距创下历史新高，标普500成分股之间的平均相关性处于自2005年以来的最低1%百分位。个股走势分化剧烈，在指数层面相互抵消。
来源：Man Numeric，MSCI Barra，截至2026年8月26日。
加载此信息图时出现问题？- 请点击此处
MacroScope的风格因子配置仍以动量为主导，偏向于已经上涨的股票，但我们也注意到质量导向因子，包括投资质量、盈利质量和盈利能力，有所上升。
该模型还做空波动性最大和流动性最差的股票，避免了在拥挤交易平仓时往往损失最大的风险敞口。按行业划分，该模型看好化肥、生物技术和石油勘探，而对抵押贷款机构、半导体设备制造商和芯片公司最为看空。
来源：Man Numeric，MSCI Barra，截至2026年8月26日。
加载此信息图时出现问题？- 请点击此处
这引出了最终的时机问题。
只有算命先生才能准确预测市场何时会反映现实。正如1997和2013所证明的那样，这些动态可能会持续相当长一段时间。历史表明，这种特殊的自满情绪一旦情绪最终转变，可能成为重大动荡的温床。所有数据均来自彭博，除非另有说明。
作者：Valerie Xiang，Man Numeric投资组合经理。[1] 过往业绩不代表未来表现
如需进一步澄清此处出现的术语，请访问我们的术语表页面。
Is the current market turbulence echoing 1997 and 2013?
It’s back to school and we have returned to our desks only to find the same unresolved tensions we left behind. No shiny new pencil cases. Instead, geopolitical and macroeconomic risks continue to rise, driving up global bond yields while equity markets still sit near record highs, propped up by tech stocks masking weakness elsewhere. And the same underlying question of when it’s all going to unravel.
While unfortunately we don’t have a crystal ball, it might help to take a look back at history. Our proprietary macro timing model MacroScope which examines past market regimes and looks for patterns that can help us read present conditions shows the current landscape bears the closest resemblance to May 1997. That was five weeks before Thailand devalued its currency and triggered the Asian financial crisis.
February 2013, three months before the Fed taper tantrum, comes second. While today’s conditions are not an exact replica of those periods, what they share with September 2026 is that the prevailing risks were widely known but not yet disruptive, allowing investors to largely look past them.
The emerging stress in Asia was initially treated as a regional and containable problem. The US economy was enjoying a 'Goldilocks' period with a rare combination of low unemployment, contained inflation and improving productivity. That crisis did not reach Wall Street until late October, when the Dow plunged 7.18% and circuit breakers halted trading for the first time.
The S&P 500 still ended 1997 up 31%, but the initial disruption triggered a wider global systematic shock which endured throughout 1998.[1]
The 2013 environment makes a similar point from a different angle. Everyone knew the Federal Reserve’s money printer would eventually be switched off, yet markets climbed anyway. The central bank had spent years buying bonds, to hold borrowing costs down and nudge investors out of safe assets and into risky ones. Growth was weak and spending cuts were about to bite, meaning when the reckoning arrived in May, no one should have really been surprised.
So, how useful is that history lesson for today? Our current stance leans toward cautious pessimism rather than bearishness. None of the obvious candidates, whether the trillions going into artificial intelligence, the war in Iran or the state of US public finances, looks likely on its own to trigger an imminent break.
A fiscally constrained US government, a Fed we believe is on the verge of raising rates, and valuations still priced for a fairly benign outcome all combine to leave the market with very little margin for error. These factors are intertwined and generally reinforce each other, suggesting it would not take much of a shock to expose the underlying fragility.
It might offer some comfort that the model cites August 2011, when the US lost its triple-A credit rating and the Eurozone debt crisis threatened the single currency bloc, as the least similar period to our current environment.
At the time of writing, Wall Street's fear gauge, the VIX, held close to its year-low of 14.2 on 14 August, marking its lowest point this year. VIX futures expiring in October had nudged up to 19 for October and 19.7 for November, anticipating volatility in the run up to the US mid-term elections. However, that’s still a good clip from the psychological threshold of 30 indicating severe market stress.
On 9 July, the gap between the volatility investors expected from the index and the volatility they expected from the companies inside it was the widest on record, and average correlation between S&P members is in the bottom 1% of readings since 2005. Shares are moving sharply in opposite directions, which cancels out at index level.
Source: Man Numeric, MSCI Barra as at 26 August 2026.
Problems loading this infographic? - Please click here
MacroScope’s style factor positioning remains dominant on Momentum and favouring shares that have already been going up, but we have also noted a nudge up for Quality-oriented factors, covering investment quality, earnings quality and profitability.
The model is also short the most volatile shares and the hardest ones to sell quickly, avoiding the exposures that tend to hurt most when a crowded trade unwinds. By industry it favours fertilisers, biotechnology and oil exploration, while turning most negative on mortgage lenders, semiconductor equipment makers and chip firms.
Source: Man Numeric, MSCI Barra as at 26 August 2026.
Problems loading this infographic? - Please click here
This brings us to the ultimate question of timing.
Only a fortune teller can predict exactly when the market will price in reality. As both 1997 and 2013 demonstrated, these dynamics can rumble on for quite some time. History shows that this specific brand of complacency can be the seedbed for major disruption once sentiment eventually turns. All data Bloomberg, unless otherwise stated.
Author: Valerie Xiang, a Portfolio Manager at Man Numeric. [1] Past performance is not indicative of future results
For further clarification on the terms which appear here, please visit our Glossary page.
正在载入文档……