Pacific Rift

Japanese equities have never been this cheap relative to the US.  Why does it matter?

By: Nick Schmitz

The gap between stock valuations in the United States and Japan has reached all-time highs after nearly a decade of growing divergence. In one country, stocks trade at 32x P/E, while across the Pacific, stocks trade in the teens. A decade ago, stocks traded at virtually the same multiples in both countries.

Figure 1: Trailing Decade of US vs Japanese Equity Valuations

Source: Ken French

In one country, a handful of mega-cap growth stocks have soared to new heights, accounting for an unprecedented share of global market cap. In the other, macroeconomic uncertainty and fears of secular stagnation have led to extremely high discounts.

Investment analysts quoted in the Wall Street Journal’s year-end review at the end of the decade acknowledged the “astronomical” valuations but argued that these valuations were justified by the “consensus emerging from [the year-end] forecasts.” The analysts argued that valuations were “chronically higher” in one market than the other, that the “outlook is extremely good” and that, after a very strong year, “we’re looking for another good year” of growth and rising valuations.

But the academics who focus on studying valuation and efficient markets had a different take. Ken French, famed Chicago School economist and longtime peer of efficient markets theorist Eugene Fama, analyzed the data between the two countries, trying to discern if the relative valuations are justified by the usual explanations: relative growth expectations, accounting differences, interest rate differentials, etc. But none of these “appear large enough to fully account for the change in stock prices,” according to Professor French. Markets, to French, appeared mispriced—an unusual statement from a research partner of Eugene Fama.

So who’s right? Investors or academics? It would be impossible to say were we describing today’s market conditions. But because Ken French wrote that in 1989 about Japan, not in 2020 about the US, and because Figure 1 shows US and Japanese valuations in the 1980s, we know exactly who was right. The quote and the data come from Ken French’s 1989 NBER working paper “Are Japanese Stock Prices Too High” (finally picked up in 1991 by the Journal of Financial Economics under the title “Were Japanese Stock Prices Too High”).

As it turned out, Ken French was right. Japanese stock prices were too high in 1989. Relative realized returns did not come anywhere near the expected returns implied by the relative prices. And buying the most expensive stocks in the most expensive and best performing market globally turned out to be one of the worst investment strategies ever, even though investors had compounded almost three times as much investing in Japan versus the US for the decade up until 1990.

Figure 2: Returns on Japan’s TOPIX index vs the S&P 500 from 1980 to 2000

Source: Capital IQ

What’s interesting about this anecdote, and the eerie prescience of Ken French’s paper, is that the valuation chart of the Topix versus the S&P 500 today looks almost exactly like 1989 with the labels reversed. Hindsight is 20/20, but 2020 valuations look like a reverse 1989.

Figure 3: Last Decade of US vs Japanese Equity Valuations (in 2020)

Source: Capital IQ

To quants, 2020 looks like a reverse déjà vu of 1989 across the Pacific rift. Today, the six FANMAG stocks (Facebook, Amazon, Netflix, Microsoft, Apple and Google) now exceed the capitalization of every stock market in the world except the US and Japan. In 1988, Japan had eight of the world’s top ten largest companies by market cap, with only Exxon making the list for the US. 

In 1990, when the US traded at ~40% of the price of Japanese stocks (the opposite of today), you earned a significant premium owning the cheaper country for the next decade. The US earned about a 20% annualized premium versus Japan for the next decade. 

In 2000, after growth stocks had run up, owning the cheapest stocks in either market earned a similar decade-long premium.

In 2010, owning anything worked. US large growth just happened to work really well, as the relative price of growth stocks approached the 97th percentile of market history in 2020 like in the run up to the 2000 bubble.
 
Figure 4: Annualized Factor Portfolio Returns by Decade – US and Japan

Source: Ken French Library. 5x5 size and book-to-market portfolios.

So going forward in 2020, it makes a lot of sense to us to invest in the cheapest stocks in the cheapest market for the next decade.

And we believe that these historical contrarian outcomes generally make sense, given that we have a hard enough time predicting next week’s coronavirus stats, or the Michigan primaries the day before, let alone next decade’s sum-total of macroeconomic and geopolitical outcomes. In such ridiculously uncertain environments, buying exposure to what is roughly equivalent from an accounting perspective, yet on sale today, is far more rational in our view.