03-16-2026, 04:11 PM
(This post was last modified: 03-16-2026, 04:12 PM by Observant1.)
testing...
Introduction
1 Using MSCI data, the 35-year outperformance to December 2024 was 4.7.% p.a. Relative valuation change contributed 3.8%, real
EPS growth edge 1.1%, dividend yield differential -0.6%, and real interest rate differential -0.3%. Exhibit 4 below shows similar
decompositions for other windows and also for dividend-based decompositions with Dimson-Marsh-Staunton data.
2 Of course, this is not a new debate. The pro-diversification literature includes Asness-Israel-Liew (2011) who show global
diversification helps over long horizons, and Dimson-Marsh-Staunton (2021) who explore the role of luck, with the US being on the
winning side of world wars and the cold war and its related economic competition, and having no wars on its own soil, revolutions or
hyperinflations. Asness-Ilmanen-Villalon (2023) find that most of US outperformance since 1990 reflects relative CAPE valuation
increase.
Part 2 of this series cuts straight to a very
topical investment decision: the allocation
between US equities and the rest of the world,
mainly focusing on developed markets.
US equities outperformed other markets in the
1990s, and the 15 years from 2009 to 2024. The
prolonged latter episode reinforced a belief
in US exceptionalism (tied to entrepreneurial
culture, market-friendly institutions, etc.) and
led many investors (especially home-biased
US ones) to ask whether they should bother
with international diversification when it
seemed to be such a return drag. Big names
like Warren Buffett and John Bogle have been
proud US-only proponents, and this view has
paid off.
There are several counterarguments. We show
that US outperformance since 1990 primarily
reflects richening relative valuations.1 By
the end of 2024, relative valuations were at a
historically extreme level, and we argue that
some mean reversion is a sounder assumption
than extrapolation of further richening.2
Moreover, while many investors do not
have first-hand memories of the pre-GFC
investment landscape, they should know that
the US has underperformed the rest of the
world for extended periods, for example the
decades of 2000s, 1980s, and 1970s. That
said, the US has enjoyed an average return
and growth edge over the past 50 or 100+
years—though not as large as implied by
recent valuations. We suspect that the recent
high valuations and predicted abnormal
growth edge partly reflect investors mistaking
the richening-driven return outperformance
for growth-driven outperformance (which
would be more reasonable to extrapolate).
We acknowledge that in the past decade,
valuation-based capital market assumptions
(CMAs) have given too pessimistic forecasts on
absolute and relative US market performance.
But we also show that over longer histories
such CMAs have been more often right than
wrong, while rearview-mirror expectations
(extrapolating the past decade’s relative
performance) have hurt investors. An
investor’s current view on US versus the rest
is a Rorschach test on whether they care more
about one recent observation or longer-term
statistical evidence.
We also briefly discuss diversification and
luck arguments, and the implications of US
“Magnificent Seven” and tech edges and
America First economic policies.
Introduction
1 Using MSCI data, the 35-year outperformance to December 2024 was 4.7.% p.a. Relative valuation change contributed 3.8%, real
EPS growth edge 1.1%, dividend yield differential -0.6%, and real interest rate differential -0.3%. Exhibit 4 below shows similar
decompositions for other windows and also for dividend-based decompositions with Dimson-Marsh-Staunton data.
2 Of course, this is not a new debate. The pro-diversification literature includes Asness-Israel-Liew (2011) who show global
diversification helps over long horizons, and Dimson-Marsh-Staunton (2021) who explore the role of luck, with the US being on the
winning side of world wars and the cold war and its related economic competition, and having no wars on its own soil, revolutions or
hyperinflations. Asness-Ilmanen-Villalon (2023) find that most of US outperformance since 1990 reflects relative CAPE valuation
increase.
Part 2 of this series cuts straight to a very
topical investment decision: the allocation
between US equities and the rest of the world,
mainly focusing on developed markets.
US equities outperformed other markets in the
1990s, and the 15 years from 2009 to 2024. The
prolonged latter episode reinforced a belief
in US exceptionalism (tied to entrepreneurial
culture, market-friendly institutions, etc.) and
led many investors (especially home-biased
US ones) to ask whether they should bother
with international diversification when it
seemed to be such a return drag. Big names
like Warren Buffett and John Bogle have been
proud US-only proponents, and this view has
paid off.
There are several counterarguments. We show
that US outperformance since 1990 primarily
reflects richening relative valuations.1 By
the end of 2024, relative valuations were at a
historically extreme level, and we argue that
some mean reversion is a sounder assumption
than extrapolation of further richening.2
Moreover, while many investors do not
have first-hand memories of the pre-GFC
investment landscape, they should know that
the US has underperformed the rest of the
world for extended periods, for example the
decades of 2000s, 1980s, and 1970s. That
said, the US has enjoyed an average return
and growth edge over the past 50 or 100+
years—though not as large as implied by
recent valuations. We suspect that the recent
high valuations and predicted abnormal
growth edge partly reflect investors mistaking
the richening-driven return outperformance
for growth-driven outperformance (which
would be more reasonable to extrapolate).
We acknowledge that in the past decade,
valuation-based capital market assumptions
(CMAs) have given too pessimistic forecasts on
absolute and relative US market performance.
But we also show that over longer histories
such CMAs have been more often right than
wrong, while rearview-mirror expectations
(extrapolating the past decade’s relative
performance) have hurt investors. An
investor’s current view on US versus the rest
is a Rorschach test on whether they care more
about one recent observation or longer-term
statistical evidence.
We also briefly discuss diversification and
luck arguments, and the implications of US
“Magnificent Seven” and tech edges and
America First economic policies.
