← 返回 Avalaches

在这篇 Unhedged 文章中,小型股在长期落后后,今年仍表现较好;部分超额报酬可归因于对不受欢迎的科技股权重较低、对能源曝险较高。与大型股相比,估值差距曾大幅扩大,在 11 月突破过 7.5 点市盈率价差;若只看五年图,可能误以为关系已回到正常,但实际上,在此前约 15 年,小型股估值溢价一直在主导。作者因此提出一个疑问:大型股价值股胜过成长股的现象,为何未能在小型股中同样出现。

比较 MSCI、S&P、Russell 三家机构的六个指数(小型价值与小型成长)发现,短中期差异很大:年初以来,MSCI 与 S&P 的小型成长指数表现高于小型价值指数,但在 Russell 的另一组指数则是反向。Dimensional Fund Advisors 的 Wes Crill 表示,指数建构与再平衡规则的细微差异,足以引发任意分歧——甚至可能只差一间股票是否被纳入其一,故其中很多变动像是噪音。跨十年观察时,这种偏差大体消退,三个小型价值指数的表现相当接近,显示建构差异多为短期效应。

但仍存在结构差异。Russell 的小型价值看似较强,可能来自其更高能源曝险和较低科技曝险;其成分也更宽松,单一最大成分市值最高、平均市值最高,但中位数市值最低,意味有更多极小企业。Rob Arnott 指出,年初成长与价值、以及大型与小型之间的估值裂口都接近历史极值,一旦大型成长(含 Mag 7)受冲击,均值回归就可启动;他认为小型成长其实是「轻成长+轻价值」混合,而小型价值是更深度被抛售的便宜股,唯恐惧压抑其反弹。Christian Caillaux 进一步指出,投资者仍想保有风险资产与 AI 曝险,但认为大型成长已偏贵且集中,因此转向市值较小、题材更广的市场,以替代较集中且拥挤的大型 AI 生态,支持了小型成长领先。

feea7f3a8238.png



The Unhedged article says small-cap stocks are still outperforming this year after a long period of underperformance, in part because they are less exposed to out-of-favour tech and more exposed to energy. The valuation gap versus large caps had widened sharply, reaching over 7.5 points in P/E in November. A five-year chart can be misread as a full normalisation, but in the prior roughly 15 years small-cap valuations had usually carried a premium. So the author asks why large-cap value beating growth did not appear equally in small caps.

Across three providers—MSCІ, S&P, and Russell—the six short-listed indices (small value and small growth) showed large short- to medium-term dispersion: year-to-date, small growth led value in MSCI and S&P, while Russell showed the reverse. Wes Crill at Dimensional Fund Advisors says small differences in index rules and rebalancing can create arbitrary differences, even down to one stock being included in one index and not the other, so much of the gap is noise. Over ten years, that noise largely fades, as the three small-value indices converge in performance, suggesting index-construction effects are mostly temporary.

But structural differences still remain. Russell’s stronger small-value results appear linked to higher energy exposure and lower tech exposure, and to a looser inclusion set: it has the largest single constituent by market value, the highest average market cap, and the lowest median cap, implying a wider range with many very small names. Rob Arnott says the spread between growth and value, and between large and small caps, was near historical extremes, so mean reversion can occur once large growth—especially the Mag 7—is hit by shocks. He argues small growth is really a mix of mild growth and mild value, while small value is deeply beaten-up and cheap, held down mainly by fear. Christian Caillaux adds that investors still want AI exposure and risk assets, but prefer small caps because the universe is broader and less concentrated than large-cap AI themes, which supports small-growth leadership.
2026-04-15 (Wednesday) · 2e6bbc0238d666b6e9c4a92d910deefb94ed66d7