A comparison of long-term trends in the US and Australian stock markets

By Robert Vagg

Much can be learned from the detailed history of a stock market that may be used to evaluate its current position and likely future direction. The field of technical analysis is based on the recognition of such definable trends and historically repetitive patterns. This article compares in detail the long-term behaviour of the US and Australian stock markets, as represented by the S&P500 Composite and the All Ordinaries Price Index, respectively, or their index precursors. Much of this analysis is based on a generic model developed by the author and described for the Australian market in the May 2009 issue of Investors’ Voice.¹

US-Australian Market Connections

The US stock market is important to Australian investors for many reasons, both fundamental and technical. Apart from those with direct holdings, investors in managed funds that have an international component have an indirect holding in the US market. So also do the many Australians who are members of large retail or industry super funds, since local markets are of insufficient size to absorb the amount of investment capital involved.

The correlation between movements in the US stock market with those of our own goes well beyond their obvious daily oscillations. As part of a globalised financial system, the two markets are closely linked. In the short term, one key reason for a correlation is the constant relative re-balancing between local and international components of equities holdings required of mutual or managed funds in order to satisfy trust deed guidelines. Thus, if the US market falls overnight the local market then is sold in order to maintain those prescribed relative weightings. This process does not involve recourse to fundamentals.

Reference to the back pages of company annual reports reveals the familiar names of large international investment houses to be commonly listed among the top twenty shareholdings of Australia’s major public companies. The latter are less Australian owned than is generally thought. To US-based mutual and hedge funds, Australia forms part of their international holdings, which must be sold down when their investors place redemptions for cash. For this reason, as was manifest in late 2008, we have a contagious one-way exposure to the prevailing sentiment of the US investor. Again, in that particular case, little regard was paid to local fundamentals.

The Primary Trends

A comparison of the long-term behaviour of the two stock markets is portrayed in Figure 1. This shows Australia’s price index, the All Ordinaries, to have significantly outperformed its US counterpart throughout its history. This outperformance would be even greater if corresponding accumulation indices were to be displayed, given the Australian market’s consistently higher dividend returns. The comparison displays an important feature. The plots of both indices show upward curvature over time, demonstrating the gradually changing nature of their underlying economies from cyclical to higher growth. This indicates a consistently increasing annual growth rate over time for both markets, which is particularly pronounced in the US market from the middle of the twentieth century onwards (i.e. after WW2).

Figure 1. Long-term Views of the All Ordinaries and S&P500 Price Indices

When plotted on a log scale, as here, a constant rate of compounding growth would appear as a straight-line trend. However, trendlines fitted to the indices in Figure 1 are clearly not straight. Rather, they demonstrate third-order compound growth, meaning that the annual growth rates also have increased in a compounding manner over time. It is common to see in the financial media similar long-term charts with straight lines incorrectly fitted to represent trend growth. Such interpretations are misleading, in that they fail to take into account the changing growth characteristics of these markets, and of their underlying economies.

The trendlines displayed in Figure 1 represent the Primary Trend that is fundamental to each respective market. The value calculated for the Primary Trend at any time would represent the Primary Mean (long-term fair value) for the market at that date.¹ It is clear that deviations from these central Primary Trends are more pronounced in the US market, indicating higher levels of volatility (and therefore higher risk). In this long-term context, the recent GFC appears as a normal divergence on these index charts. These volatility characteristics are examined in further detail below.

Figure 2 displays the changing annual growth rates for these markets calculated from their respective Primary Trends. These plots demonstrate the different growth characteristics of the two markets, as well as the compounding nature of their growth rates. The Australian market began its history near the end of the 19th century growing at an annual rate of 4.0%, which has increased steadily to now average 9.4% pa. By comparison, the US market entered the 20th century growing at a rate of only 1.4% pa, but has displayed higher growth since the later part of that century.

An interesting issue arises from the compounding nature of these growth rates. Since price/earnings (P/E) ratios contain a component that allows for perceived growth in future earnings, this component should increase along with increasing growth rates. As a result, market P/E ratios should slowly expand going forward rather than fluctuate around a static average, and hence it would be arithmetically incorrect to quote a “long-term average market P/E”.

Figure 2.  Stock Market Trend Annual Growth Rates

The two markets are compared in greater detail in Figures 3 and 4. Here the index values are plotted using the price range for each year since 1875.  The availability of values calculated for the two Primary Trends allows a comparison of indicated fair value for the two markets at any time. Mean values calculated for 1 January 2010, for example, are All Ordinaries 6376 and S&P500 2037, whereas corresponding values for 1 January 2015 are 10118 and 3503, respectively. Such an exercise also affords a method of estimating the degree of over- or undervaluation prevailing in those markets. Examples are presented in the diagrams for some infamous market extremes.

Figure 3.  The All Ordinaries Price Index Primary Model

Figure 4.  The S&P500 Composite Price Index Primary Model

Volatility

As well as the Primary Trend for each index, statistically-derived 90% confidence limits are shown as lines running parallel to the central trendlines to form channels that quantify their volatilities.^{1 , 2} Comparison of the relative widths of the two channels demonstrates the higher volatility of the US market. Its channel limits calculate at 55% above and 35% below the central trend position. The corresponding channel for the Australian index is narrower and better defined, with respective limits at +33% and ‑27%. The index values are seen to meander consistently within these two channels, rarely wandering beyond their limits. In this respect, it is interesting to note that in their volatility extremes there is little correlation between the two markets.

Percentage movements in the indices relative to their central trendlines are illustrated in Figures 5 and 6. These allow graphical comparison of volatility patterns that characterise the two markets and would permit an easier estimation of each market’s relative valuation at any time.

Figure 5.  All Ordinaries Volatility Relative to Primary Mean

Figure 6.  S&P500 Volatility Relative to Primary Mean

Excessive volatility in the Australian market has occurred on only three occasions – 1974, 1987 and 2009. The US market achieved its abnormal extremes in 1929, 1932, 2000 and 2009. Of particular interest is the fact that there is only a minor correlation between the volatilities of the two markets in both 1974 and 1987. The recent GFC, therefore, represents a unique common feature. Comparison of Figures 3–6 might suggest why US markets have been consistently compared with the 1929 Crash during the recent crisis, whereas the Australian market more often has been compared with the falls of 1974. Similar levels of undervaluation were achieved in each case.

The excessive volatility of the US market from 1929 through the Great Depression has been referred to frequently, with the 1929 peak not being achieved again until 25 years later. There have been occasional suggestions recently that the US market now might follow a similar course. Close inspection of Figure 4 provides information on that period. From the bottom reached in 1932 the market in fact demonstrated above-average growth for the next three decades. In particular, the five-year period through 1932-37 was a bull market. That the 1929 peak was not reached again until 1954 may be ascribed to the gross overvaluation (of 178%) created by the bubble formed at the end of the ‘roaring twenties’ decade rather than slow growth from 1932 onwards. The Great Depression period in Australia also witnessed a six-year bull market, although the preceding peak and crash were nowhere near as extreme as in the US. The period following the 1974 low in Australia displayed the highest growth in the history of the index.

Since WW1 a volatility cycle of 10-years approximate length has been apparent in the Australian market (Figure 5), peaking near the end of each decade. This has been described previously as this market’s Secondary Cycle.¹ US volatility over the same period (Figure 6) would appear to be better described by a corresponding decade cycle superimposed upon a broader underlying 34-year cycle. The latter indicates formation of a major low in the US market around this current period.
The volatility demonstrated by the indices may be compared on a shorter timescale. Figure 7 profiles the typical year, illustrating the average positions of the two markets for any month relative to the previous one, derived from mid-range monthly values for each index since 1950. An obvious correlation exists between the two markets in this timeframe. The phase from September through November is conspicuous historically as a period of little or negative returns, although inspection of the underlying data reveals that much of this effect may be ascribed to a few well-known significant falls occurring around October. New-year optimism is apparent in the January values, perhaps due to a degree to media promotion at that time of year. Remembering that the data derives from price indices, the dips present around February/March and August/September in part may be due to shares commonly going ex-dividend around these times, a matter of particular relevance to the Australian market.

Figure 7. Average Monthly Change In Indices (1950-2009)

A note of caution is appropriate here. Diagrams similar to Figure 7 for individual markets appear quite frequently in the media. The average values portrayed here range from ca +2.5% to ‑0.5%. However, the standard deviations calculated for each month range from 3.6% to 6.7%, indicating that the errors involved are significantly greater than the average values reported. Hence little confidence should be placed on their interpretation for predictive purposes.

Currency Exchange Effects

The international holdings of Australian investors, either direct or as a component of a balanced growth fund, are very much dependent on currency exchange rate movements. As Figure 8 demonstrates, there can be a stark difference in investment performance if the S&P500 is expressed in either Australian or US dollars. An  unhedged Australian holding invested in the US since 1970, for example, would have benefited from a falling trend in the $A/$US exchange rate to achieve outperformance by September 2000 of a little over 100%. However, this advantage has been diminishing gradually throughout the current decade as a result of a reversal of this trend.

Figure 8. S&P500 Index in $A and $US (1970-2009)

Figure 9. S&P500 Index in $A and $US (2000-2009)

Figure 9 demonstrates the result of an equivalent investment made at the peak in September 2000. Represented in $A, the index has been in continuous decline, and currently exhibits 41% relative underperformance. In particular, it may be seen that the recent recovery in the US stock market since early 2009 of ca 50% will have been negated for an Australian investor by a corresponding relative depreciation of the $US. With Australia’s economic performance, and hence the $A, now expected to be more closely linked to nearby higher-growth developing markets, it is possible that this declining trend may continue for the foreseeable future.

Robert Vagg is a member of the AIA. (Contact email: rsvagg@gmail.com)

Back to top