Testing Dual Momentum in Japan

Any investment strategy you contemplate should be stress-tested for performance in bad times as well as good times. One of the best, true scenario, ongoing investor stress tests is still occurring in Japan right now.

Japan is the market that no buy-and-hold investor wants to talk about. It has been a true nightmare scenario. The average Japanese equity investor is hardly better off now than they were in 1990. A stock/bond portfolio would have experienced pitiful returns, even after factoring in currency strength.

In the past 30 years, Japanese investors have gone through two massive bear markets. One jaw dropping ~70% decline took more than a decade to hit bottom (1990 to 2003). Then, just as things started to look up, investors were hit by another ~60% crash from 2007 to 2009.

So naturally I had to see what would happen if a Japanese investor was using a local version of the Dual Momentum strategy.

Background of Japan's Economy

The Japanese nearly lost their entire economic base as a result of World War 2. Thanks to U.S. financial backing, a hard-working population, and a culture of perfection, Japan's economy quickly began to boom. By the 1980s, Japan developed into a global export powerhouse and first-world economy.

From the 1950s on, Japan's economy grew at annual rates of 8% or higher for several decades. The Japanese developed unmatched technical expertise in certain industries. The western world—still young enough to remember World War II—feared Japan may take over the U.S. as the global financial and economic centre.

After export growth slowed, domestic consumption skyrocketed in the 1980s. Property prices got so expensive in the late 1980s that the land in Japan was worth more than the entire United States. As you could imagine, much of this was propped up by massive loan growth. (Sound familiar?)

However, Japanese assets reached their peak valuations in the beginning of 1990. Since then real estate prices have dropped to a fraction of their former value. Bond yields have dropped to nothing as the government tried to stimulate the economy with negative interest rates. The stock market collapsed, led by big financial corporations and companies with high real estate holdings.

Despite the economic turmoil, Japan still has a powerful economy with many large corporations. The Japanese stock market is the second most diverse investable market in the world, following the U.S. markets.

Japan's currency, the yen, is a popular reserve currency which competes with the Euro and British pound for prominence behind the U.S. dollar. The Japanese yen has actually strengthened in value compared to other major currencies since 1990, including the U.S. dollar.

The Power of Dual Momentum

Dual Momentum is a simple timing strategy that can be used by any investor to reduce risks and achieve better investment performance. I talk about Dual Momentum in detail in the model portfolio page.

In a Dual Momentum evaluation, an investor assesses the past 12-month performance of three broad assets at the end of each month. After calculating the performance for each asset, the investor performs a momentum evaluation to determine the Dual Momentum signal for the following month.

First, the investor performs a relative momentum evaluation, comparing the performance for the two equity assets. Once the best performing equity asset has been identified, the investor performs a second absolute momentum evaluation comparing the preferred equity asset to the risk-free rate of return.

If the selected equity asset has outperformed the risk-free asset, the investor invests their entire portfolio in the lowest cost version of that equity asset. If the risk-free asset outperforms, the investor puts their entire portfolio into bonds.

For this Japanese Dual Momentum test, I used the following asset data. All assets include distributions (gross returns) and are priced in Japanese yen.

Equity Assets

  • MSCI Japan Index (1970-2017)
  • MSCI Kokusai Index (1970-1987), replaced by the MSCI ACWI ex Japan (1988-2017)

Risk-Free Asset

  • Japan Treasury Bills (1970-1979), replaced by Japanese 3 Month CD (1979-2017)

All of these assets were priced in the local currency (Japanese yen). Remember, the point of the backtest is to see the theoretical results of a Japan-based investor, not a U.S. or Canadian investor using these indices.

Japanese Dual Momentum vs. Buy & Hold

Dual Momentum (Japan) Results

This test is a full return profile for Dual Momentum since 1970—the earliest point where global broad stock data has become available. In the earliest points of this graph, Japan's economy could still be considered a developing market with the higher volatility that is characteristic of less developed markets.

The returns that Dual Momentum would have generated for a Japanese investor are nothing short of phenomenal. Japan was a true wealth building economy!

But as you can see, buy-and-hold would have worked fantastically as well. Until 1990 that is...

Source: TheRichMoose.com, MSCI, FRED

Dual Momentum would have provided the following results for Japanese investors.

Compound Annual Return:  +10.46% (49 years)
Largest Annual Gain:  +117.05% (1972)
Largest Annual Drawdown:  -21.18% (1973)
Peak to Trough Drawdown:  -29.96% (1987)

In comparison, the MSCI Japan index (in yen) returned the following results.

Compound Annual Return:  +6.28% (49 years)
Largest Annual Gain:  +117.05% (1972)
Largest Annual Drawdown:  -42.48% (2008)
Peak to Trough Drawdown:  -67.62% (1990-2003)

After the Bubble (Post 1989)

Of course it is pointless to test the robustness of Dual Momentum if we don't take a closer look at the past three decades where Japanese investors suffered from a massive correction.

As the graph shows, Dual Momentum really shines in turbulent times. While a buy-and-hold investor would have lost money, the momentum investor would have increased their wealth 10x in the same time period.

Source: TheRichMoose.com, MSCI.com, FRED

Since the end of 1989, the peak of the Japanese bubble, Dual Momentum provided the following results for Japanese investors.

Compound Annual Return:  +7.47% (29 years)
Largest Annual Gain:  +54.80% (2013)
Largest Annual Drawdown:  -13.67% (2000)
Peak to Trough Drawdown:  -22.92% (2015-2016)

In comparison, the MSCI Japan index (in yen) returned the following results after the bubble.

Compound Annual Return:  -0.76% (29 years)
Largest Annual Gain:  +54.80% (2013)
Largest Annual Drawdown:  -42.48% (2008)
Peak to Trough Drawdown:  -67.62% (1990-2003)


The Japanese case study provides an interesting insight into how Dual Momentum works across a broad range of market conditions. Dual Momentum is a robust investment strategy that performs well, even when the local markets experience big setbacks.

During the Japanese bull market preceding 1990, the momentum investor would have actually lagged the buy-and-hold investor by a small margin. An all-equity buy-and-hold investor would have grown their portfolio at an astounding compound return of 17.34% per year.

After the crisis began to unfold, Dual Momentum began to show its true strength. A compound return of 7.47% per year is an extremely good result over a three decade period, especially compared to a negative annual return for a buy-and-hold investor.

Comments & Questions

All comments are moderated before being posted for public viewing. Please don't send in multiple comments if yours doesn't appear right away. It can take up to 24 hours before comments are posted.

Comments containing links or "trolling" will not be posted. Comments with profane language or those which reveal personal information will be edited by moderator.

2 Replies to “Testing Dual Momentum in Japan”

  1. Thanks Daren. I wonder how much of the time post-89 was spent in the ex-Japan fund – presumably almost all of it? So TADM vs buy and hold of an all-world index fund would be an interesting comparison. You would think that an ideal “mix” of assets for DM would spend a more equal time in each asset, but perhaps that’s not so important after all.

    1. Daren (Editor) says:

      You would think Japan would be left out post 1989, but surprisingly the signal is in Japan more than 25% of the time.
      If you only analyze two asset classes (ACWI vs T-bills), it would be a Absolute Stock Momentum analysis. I did a simple backtest for this post: http://therichmoose.com/post20180209/
      The results were good, but not great. Definitely beat buy-and-hold, but didn’t come close to TADM over the same time period.

Leave a Reply

seven × = fifty six