We're going to look back at the boom years in Japan during the 1980s when the Nikkei225 rose from around 8000 to almost 40000 in a space of 7 years up until the end of 1990. Today the Nikkei225 has still not fully recovered and sits around 8500:
Historically (and even today) dividend yields in Japan have been very low. Back in the 1950-70s Japan was seen as an emerging market and its GDP growth was very high (not too dissimilar to China today). By the 1980s the growth had slowed down but its stock market continued to boom.
The question I asked myself was: Did Japan look a good investment during the 1980s, and would an investor have been caught in the bear market that followed in the 1990s.
If you remember from previous postings (core funds) the valuation method is based on a growth rate (derived from real GDP growth estimation (50% of this) plus an inflation estimation to give a nominal growth rate), an adjustment rate (we assumed that long term fair dividend yield is 4.5% so funds paying less than this would have an adjustment factor built into the expected return to compensate for their "overvaluation"), and the current dividend yield.
During the period from 1982 to 1992 the dividend yield ranged from a high of just over 1.6% to just under 0.4% during the high of the boom. Clearly Japanese stocks never looked cheap during this period.
What about growth? As I mentioned earlier most the high growth in Japan happened in the period after WWII in the 50s, 60s and 70s. During the 1980s real GDP growth ranged between 3 and 5% a year (hardly China or India levels), and the maximum of just under 7% in the late 1980s.
The model would have classed Japan as overvalued throughout the period in question. In fact the expected returns were negative for the most part (realised returns were also negative):
The above chart shows the expected and realised returns based on buying the Nikkei225 and holing for 20 years. As the model is quite conservative you can see that the realised returns were usually higher than the expected values (in fact they were always higher as the data ignores dividends. Finding total return data for the Nikkei225 is almost impossible. As dividends on the Nikkei225 have always been low the actual results wouldn't vary by too much). As the expected returns were so low you might have even tried to short it (I wouldn't have recommended that though). So the model steered us away from that bubble, you might feel it was a shame that we didn't get at least some of the early growth. That's true but as the Nikkei225 is still at the levels it was 30 years ago and with a current dividend of 2.4% (earnings have risen to increase the dividend yield) it still doesn't look that cheap today.
Data used from The Japanese Asset Price Bubble; Barsky (University of Michigan)