Index Investing5 分で読めます

The History of the Value Factor

Do stocks that look cheap really deliver higher returns over the long run? This belief, held to be true for decades, has been badly shaken over the past 10-odd years.

The Value Factor and the Fama-French Model

The value factor refers to the 'tendency of stocks trading relatively cheaply — like low-PBR, low-PER — to deliver higher returns over the long run than expensive growth stocks.'

What cemented this concept in academia was Eugene Fama and Kenneth French. In a 1992 paper the two presented empirical evidence that size and value explain returns, and in a 1993 paper they formalized this into the 'three-factor model.'

The three factors are market, size (SMB), and value (HML). Here, HML (High Minus Low) is the value factor, meaning 'the return of cheap stocks (high book-to-market) minus the return of expensive stocks (low book-to-market).'

How Large Was the Historical Premium

Historically, the HML (value) premium is tallied at roughly 3–5% per year. However, this value varies greatly by sample period.

For example, if you take a very long period since 1926, one tally puts the global value premium in the low 3% range per year (about 3.3%).

What matters is that this 'average' does not show up steadily every year. Some decades are strongly positive and some are strongly negative — the variance is very large.

Figures like 3–5% per year and about 3.3% since 1926 are approximate ranges cross-referenced from tallies by Alpha Architect, Dimensional, and others. They vary with the indices, periods, and regions used, so do not take them as absolute values.

The 'Value Winter' and the 2022 Rebound

The most famous event in the value factor's history is the long slump of the 2010s, the so-called 'value winter.'

As low rates and a tech-led market continued, growth stocks got far ahead. According to one tally, over roughly 14 years from 2007 to 2020, global value stocks lagged growth stocks by about 5.7 percentage points per year on average. In particular, from early 2017 to August 2020, the U.S. small-cap value research index returned a total of -13%, while the small-cap growth index returned +71% — an extreme gap. During this period, many people declared 'the value premium is dead.'

But around 2022 the flow changed. With inflation and rising rates, companies with solid earnings and cash flow were favored, and value stocks rebounded. In one tally, from the start of the recovery through June 2022, value beat growth by an annualized roughly 9.9 percentage points.

The 'value winter' figures (2007–2020 lag of about 5.7pp per year; 2017–2020 small-cap value -13% vs. small-cap growth +71%; 2022 rebound of about 9.9pp) are values confirmed from multiple sources such as Alpha Architect, Dimensional, and Morningstar. They vary by tallying agency and period.

The Premium Can Also Disappear

The lesson from the value factor's history is clear. A factor premium is 'something that may exist on average yet fail to appear for a very long time (10+ years).'

A strategy that becomes widely known can weaken as many people follow it, and if the market structure changes, the possibility that it disappears entirely cannot be ruled out. 'It existed in the past' does not guarantee 'it will exist in the future.'

So the value factor is safer understood not as a 'sure excess-return formula,' but as 'one tilt that requires the resolve to endure for a long time, and whose future is uncertain.'

よくある質問

Q. Do cheap stocks always beat expensive stocks?

No. On a long-term average there were stretches where value stocks were ahead, but there were also periods — like the 2010s — where they lagged growth stocks badly for more than a decade. Closer to the truth is 'the results diverge greatly by period,' not 'value always wins.'

Q. If I invest in the value factor, will I earn excess returns?

It can't be guaranteed. The historical premium is only an average; you may have to endure a very long slump, and there is no basis that it will persist in the future. This article does not recommend any particular factor or product, and past performance does not guarantee the future.