Japan: The ‘Bubble’ That Never Burst?

Russell Napier, author of 'Anatomy of the Bear' puts Japan's equity market in its historical perspective

Russell Napier, Consultant, CLSA, Asia Pacific

It seems to be a 'truth universally recognised' that Japanese equities were over-valued in the late Eighties 'bubble' market. Of course the very use of the word 'bubble' to describe the condition of equities in that period also implies that the 'bubble' burst in the near 80% decline in the Nikkei from its peak to its low in 2003. The language provides a convenient schemata which is wrong and dangerous. A study of over 100 years of equity market valuation suggests that Japanese equities were, in many ways, as distinctively priced in 2003 as they were in 1989. Of course some investors, transposing absolute valuations levels from other jurisdictions, would prefer the term 'wrongly priced' as opposed to the term 'distinctively priced'. Whether one chooses the word 'wrongly' priced or 'distinctively' priced will play a huge role in whether one misses one of the greatest investment opportunities of our generation. The investor who goes beyond the transposing of valuation criteria will not be catapulted out of Japanese equities and should thus participate in a rise in the Nikkei which could easily exceed 700% from the April 2003 low.

Anatomy of the Bear

For the past few years I have been researching and writing a book called Anatomy of the Bear: Lessons from Wall Street's four great bottoms. The prime aim of the book was to search through bear-market period copies of The Wall Street Journal to discover whether anybody did indeed "ring a bell at the bottom of the market". Well, there is no bell, but there is a combination of events which occur and significantly assist ident-ification of a bear-market bottom. If there is one strategictake-away from the book, it is that the key cause of major bear markets is a disturbance to the general price level. It is inflation followed by deflation, or the real threat of deflation, which reduces equities to very cheap levels and a return to price stability which has launched the greatest bull markets of the past 100 years.

In three of the four periods studied this price disturbance was inflation followed by deflation. Only when price stability returned in 1921, 1949 and 1982 did the Dow Jones Industrial Average (DJIA) bottom. While only the first two episodes saw deflation in the general price level, all three episodes were marked by major declines in commodity prices. The famous bear market of 1932 also saw a deflation in the general price level but on this occasion it was preceded by price stability rather than inflation. In all four cases a return to price stability has augured the birth of a major bull market. Why has this transition been so important and what does it mean for Japan?

The importance of price stabilisation to equity prices is not difficult to see. Assessing the fairvalue of equities is a very risky business in a deflationary environment. When prices are declining there are negative implications for corporate revenues and also nasty geared impacts on earnings. History shows that deflation can also produce disruptions to the financial system, which force government bond yields higher despite deflation. Should deflation emergethere is a risk, given that nominal interest rates cannot fall below zero, that very high real rates of interest may come to pass. With such uncertainty regarding the risk-free rate and future cash flows, risk premiums have to rise and the price of equities decline. This process goes into reverse when the general price level stabilises.

Crucially the historical analysis makes clear the absolute key importance of price stability to this bull market in Japanese equities. While many causes are given for the markets' strong rise since 2003, the analysis in Anatomy of the Bear suggests that the rise in the Nikkei 225 is almost all due to the return of price stability to Japan. Investors who do not believe that deflation is dead in Japan should not be holding Japanese equities. Regardless of whatever other positives you may see on the horizon, these are irrelevant if deflation returns. There are other important implications for the future price of Japanese equities for those who believe that inflation has returned.

The other key take-away from the historical analysis is that the rebound in the Nikkei from April 2003 to the current level is nothing exceptional. Figure 1 looks at the percentage rise in the DJIA following those four key periods in history when price stability was evident and the bear market became a bull market.

The pace and magnitude of the rebound in the Nikkei has outstripped that of the average 81% rise in the DJIA following transitions to price stability. The Nikkei's 33-month bull market had pushed equity prices 108% higher. However, the key feature of Figure 1 is that the performance of the Nikkei is not dramatically different from similar turnarounds in US equities. The current level of the Nikkei has been reached about seven months earlier than one would expect after taking the average rebound of the DJIA over all four periods.

Figure 1 – Progress of DJIA from bottom of major bear markets

  Months to rise 108% Rise within 33 months (%)
From 1921 47 +40
From 1932 10 +154
From 1949 60 +64
From 1982 42 +67
Average 40 +81

Source: Dow Jones & Co

The good news for investorsis that none of the bull markets which followed the return of price stability in the US lasted just three years or produced a rise in the index of just 108%. Figure 2 shows the duration of the bull market in months and the entire percentage rise of the DJIA over the period.

Figure 2 – Duration and magnitude of DJIA rises following price stabilisation

  Duration in months % Rise
1921 97 +491
1932 56 +372
1949 234 +509
1982 208 +1408
Average 149 +695

Source: Dow Jones & Co

Perhaps a good generalisation, based on the data, would be that the 108% rise in the Nikkei since April 2003 can be considered small beer in the story of bull markets which follow the return of price stability. The four charts adjacent look at the 10-year period following on from the deflation-to-inflation transition in US history which began in 1921, 1932, 1949 and 1982. The Topix Index since April 2003 is rebased to coincide with the bottom of those four great bull markets of US history.

Figure 3 - The DJIA 1921 - 1931 and the Topix April 2003 - January 2006, rebased to 100
Figure 4 - The DJIA 1932 - 1942 and the Topix April 2003 - January 2006, rebased to 100
Figure 5 - The DJIA 1949 - 1959 and the Topix April 2003 - January 2006, rebased to 100
Figure 6 - The DJIA 1982 - 1992 and the Topix April 2003 to January 2006, rebased to 100
Figure 10 - Japan's cyclically adjusted PE

The charts above suggest that the equity market is not particularly ahead of itself in its rise to reflect the transition from deflation to inflation. It also suggests that the bull market remains in its early days. Of course, over such long bull markets more than price stability was playing a role in pushing equity prices higher. Figure 7 shows the growth in earnings per share and dividend per share over the period.

Figure 7 – The role of earnings and dividends in US bull markets (S&P composite index)

  Growth in earnings (%) Growth in dividends (%)
1921-1929 +238 +97
1932-1937 +144 +25
1949-1968 +478 +373
1982-2000 +280 +140

Source: Homepage of Robert Shiller of Yale University

What is particularly interesting about the figures above is that earnings and dividend growth in these bull markets have been generally well below the rise in the price of equities. This is the case even though earnings and dividends have normally reached very depressed levels at the bottom of bear markets. A rise in the valuation of equities has clearly played an important role in the story of these bull markets. One way of ironing out the cyclicality of earnings and focusing on the role of rising valuations is to calculate the cyclically adjusted PE. The current cyclical adjustment of choice, recommended by Professor Shiller of Yale, is to use the 10-year rolling earnings per share figure.

Figure 8 – Expansion of cyclically adjusted PE during the bull market (S&P Composite Index)

  Start (x) Finish (x) % rise
1921-1929 7.4 13.9 +88
1932-1937 4.7 20.1 +327
1949-1968 11.7 25.1 +115
1982-2000 9.9 48.6 +390
Average     +230

Source: Homepage of Robert Shiller of Yale University

These rises in the cyclically adjusted PE (Figure 8) contrast with the rises in cyclically adjusted earnings over the course of these four bull markets.

In general, the bulk of the return accruing to investors over the course of these bull markets is accounted for by rising valuations rather than rising earnings. Simply put, one should expect bonds to be highly valued relative to equities during deflationary episodes. This undervaluation is then unwound in periods of inflation and it is this impact, rather than any rebound in earnings, which accounts for the greatest proportion of total returns from equities.

The bubble that never burst

So what does this mean for Japan when equities did not drop to valuation levels seen in US deflationary periods? In other words, does Japan now operate in an environment were normal valuation models limit the price paid for equities or does its very high savings level lead to very high valuations when investors are forced to flee the risky bond market? There is much in the high level of the cyclically adjusted PE in Japan's deflationary era to suggest that equity valuations are headed much higher from current levels. It is this indicator which suggests that Japanese equities may be the bubble that never burst!

Figure 9 – Role of rising valuations & earnings in US bull markets (cyclically adj. S&P Comp.)

  % rise in valuations % rise in Cyclically adjusted earnings
1921-1929 +88 +12
1932-1937 +327 -15
1949-1968 +115 +256
1982-2000 +390 +168
Average +230 +105

Source: Homepage of Robert Shiller of Yale University

For the US market, it is possible to calculate cyclically adjusted earnings and thus the cyclically adjusted PE going back to 1881. In Japan the available time series for earnings is just a few decades. Thus, there is something important which the US data can tell us but the Japanese data cannot. The time series is too short to tell us anything regarding the appropriate average or even the range of the cyclically adjusted PE for Japan.

Thus, most of the value in the analysis is negated. However, we can still gauge the extent of the potential bull market by looking at the average increase in cyclically adjusted earnings and the cyclically adjusted PE one would expect in a normal US post deflation bull market. If cyclically adjusted earnings will increase by the average of 105% and the cyclically adjusted PE will increase by 230%, then the current bull market will not be complete until the Topix is valued at close to one hundred times earnings and is around the 5,700 level!

Figure 11 - The cyclically adjusted PE of the S&P Composite Index


In the US equities trade at <10x earnings in a deflationary period but in Japan around 30x!

It is important to remember that we do not have sufficient data to estimate a "normal" cyclically adjusted PE for Japan. The analysis focuses instead on the fact that Japanese equities did not reach the low valuation levels associated with deflation in the US. This exercise suggests Japanese equity valuations must have been at their lows in a deflationary environment, as they were in the US, and thus will expand from the 29x earnings level.

Japanese investors remain huddled in the safety of bonds and yet the cyclically adjusted PE of the equity market already exceeds fifty times! It will be the return of inflation which will force Japanese investors to abandon their love affair with bonds and, given the nature of institutional fund management, a significant portion of the resultant cash will be invested in Japanese equities. All the evidence from the US is that one should expect a rise in the cyclically adjusted PE to around 100X.

The great surprise of this Japanese bull market may not be that foreign investors were early to the party. It may be that they, driven by value considerations transposed from another jurisdiction, left the party just as things were livening up.