Leading Indicators of the Shanghai Composite

Includes: CDCAQ, FXI, SPY, UUP
by: Matthew Claassen

Since early 2009 many investors have focused on the trend of the Shanghai Composite as a potential leading indicator for U.S. markets. The reasoning behind this theme is simple; China appears to be a driving force in global economics, and the Shanghai Composite peaked just prior to the US markets in 2007 and troughed ahead of the US in 2009. While we do not wish to discount the importance of China’s economic growth on the world stage, there is reason to suggest the Shanghai Composite may not be a reliable leading indicator of US markets. We will detail that argument in a moment, but first; we need to understand why the Shanghai Composite seems to be doing such a poor job of leading the US and Chinese economies in 2010.

The Shanghai Composite Index peaked eight months ago, in the first week of August 2009. Assuming the historical norm of major market turning points three to nine months ahead of the economy, there should be some evidence of slowing growth within China’s current economic reports. Instead, the latest GDP reports indicate the Chinese economy grew 11.90% year over year, property prices jumped at the fastest rate since at least 2005, consumer prices rose at the fastest pace in eighteen months and lending exceeded economist’s forecasts.

Is the Shanghai Composite failing us? Not entirely. The chart below illustrates the Shanghai Composite (SSEC) in blue and an Equal Weighted Index of the Shanghai Composite (our creation) in red. Like many major market indices, the Shanghai Composite is a capitalization weighted index. The Shanghai index utilizes a Paasche weighted composite formula that divides the components into nine categories from lowest 10% of capitalization to the highest 20% of capitalization, then weights by category. As a result, the top 20% of components receive the highest weighting, with the remaining categories ranked at 10% increments. Conversely, an equal weighted index favors smaller cap issues because all issues are given the same weighting and there are a significantly greater number of smaller cap issues than large cap.

What the chart below is showing us is that while the large cap issues within the Shanghai Composite peaked eight months ago, smaller cap issues represented in the Equal Weighted Index have continued to post new rally highs, with the most recent top in early April, 2010. The Equal Weighted Index has tracked the S&P 500 much more closely than its weighted counterpart.

Shanghai Composite vs. Equal Weighted Shanghai Index

The bottom panel of the chart illustrates the relative performance of the Equal Weighted Index to its weighted counterpart the Shanghai Composite (SSEC). As would be expected, the upward trending line shows that smaller cap issues have outperformed the large cap issues since the index bottomed in 2008. I would suggest that a true downtrend in the Shanghai Composite should be preceded or accompanied by a downtrend in the relative performance line of the Equal Weighted Index vs. the Shanghai Composite. When that occurs, then the smaller cap stocks will have joined their larger cap counterparts in a downtrend, dragging the entire index lower. It is possible the April peak in the relative strength line is that top, but the relative performance trend has not yet confirmed that assumption.

The chart below illustrates Buying and Selling Volume within the Shanghai Composite. To be clear, these are not simply a measure of index volume. When calculating Buying and Selling Volume for any index we aggregate the buying and selling volume for each of the index components on a daily basis. What this chart shows is clear; average Selling Volume has risen sharply since early April and is now greater than average Buying Volume for the first time since the index began its rally in November, 2008. We do not use this indicator alone to judge whether an index should be bought or sold. But the message here is clear; the relationship between Buying Volume and Selling Volume is beginning to favor sellers.

Shanghai Composite Buying Volume vs. Selling Volume

From current levels, the Shanghai Composite is on the cusp of providing a sell signal. Further weakness in the days ahead would place the relative strength of the Equal Weighted Index in a new downtrend and push Selling Volume into a more dominant position over Buying Volume. If that should occur, we would witness the Composite move from an eight month trendless pattern to a new downtrend. Still, there is one other indicator that would best tell us if the probabilities lean toward a breakdown, or a bounce from current levels.

However, before we look at that indicator, or extrapolate a potential downtrend in the Chinese markets to a collapse in the US markets, we should at least look at the Shanghai Composite’s historical record as a leading indicator of US markets. In addition, considering that an uptrend in small cap stocks typically reverses prior to large caps; what is the divergent strength in the trend of Chinese small cap stocks telling us about their economy?

Readers of my past report Market Breadth: A sign of Health of a Symptom of Market Liquidity may find the above chart comparing performance of the Equal Weighted Index to the weighted Composite eerily familiar. The reason is that the chart is very similar to the S&P 500 vs. the small cap S&P 600 during the period from 2000 to 2002. It may be no coincidence that both periods of time represent two markets influenced by excess central bank liquidity and a growing bubble in real estate prices. Considering such, it is worth noting that although the 2000 to 2002 decline in equities was substantial, money flowing from the equity market during that period fueled the real estate bubble for several more years. There was a minor recession in 2001, but the Great Recession didn’t start until after the real estate bubble burst.

The same could unfold in today’s China. A bear market in equities may not translate into a demise of their real estate bubble. In that case, China’s economy may continue to press forward with their equity markets not entirely reflective of their economy and, in turn not necessarily reflective of the economies that depend on China for trade. Of course, Chinese regulations limiting real estate investments may also be a factor.

S&P 500 and Shanghai Composite Historical

We commented in the first paragraph that the Shanghai Composite may not be as reliable a leading indicator of US markets as many investors believe. We see the reason for this is twofold; first is the correlation of China’s equity markets and real estate bubble to the US at the turn of the century, the second reason is mechanical.

Mechanically, the Shanghai Composite lacks two features we find necessary for a consistently reliable leading or coincident indicator; historical correlation and low or equal relative volatility. The chart above overlays the Shanghai Composite (blue) and the S&P 500 (black) from 1995 to present. Other than the 2007 peak and 2008 low, this Chinese equity index has shown little correlation to US markets and no reliable ability to predict turning points. One could reasonably argue, from an historical perspective, that the apparent leading indications of the Shanghai Composite to the US market in 2007-2008 were simply coincidence.

Historical Rolling Five Day Volatility Comparison

However, even if it were assumed that China’s growing global influence were producing a more reliable correlation between the Shanghai Composite and the US equity indices, there is a problem of relative volatility. For one to trust an indicators’ trend change as predictive of a trend change in another index, it is imperative that your indicator is not more volatile than the index it is expected to predict. A significantly more volatile indicator will, by definition, introduce more noise and generate false signals. This is one reason why economists and market analysts apply moving averages to their data to reduce noise.

As an indication of relative volatility; the above chart illustrates a rolling five day percent price range of the Shanghai Composite (blue) and S&P 500 (red). During the last ninety days, the average rolling five day percent range (from low to high) of the Shanghai Index is 4.35%, or roughly 35% greater volatility than the S&P 500. As is also apparent in the above chart, pre 2009 the relative volatility was significantly greater.

Considering the low and inconsistent historical correlation along with higher volatility, there is no assurance that a breakdown in the Shanghai Composite Index would be followed by a downtrend in US markets, or that a relief rally in Shanghai Index will support US equity prices. There is a much more reliable and consistent indicator to use with the US market that we will reveal in the next public report. For now, let’s wrap this up with a look at the “one other indicator that would best tell us if the probabilities lean toward a breakdown, or a bounce from current levels” promised earlier.

Percent Above 10 Day, 50 Day and 150 Day Averages

Among the most useful measures of overbought or oversold for any index is a reading of the percent of issues above their 10 day, 50 day and 150 day moving averages. The percent of issues above their 10 day average is a very sensitive and short term indicator. Historically, when the percent of issues above their 10 day average is less than 20% a multi day rally is near at hand. Conversely, when the same indicator is above 80%, the index is likely to experience a multi-day consolidation. The same indicator applied to the percent of issues above their 50 day average and 150 day average is used slightly differently.

During bull markets, the percent of issues above their 50 day average seldom stays oversold for long. As shown in August and September of ’09 in the center panel of the above chart, when fewer than 20% of the index components are above their 50 day average it is a buying opportunity, in a bull market. In addition, bull markets seldom bring the percent of issues above their 150 day average to below 20%. Such a low reading is typically reserved for bear markets only. However, it is not uncommon for prolonged sideways consolidations, similar to the Shanghai Composite performance since August ’08, to bring this indicator very near 20%. The bottom line is that all three time frames show the Shanghai Composite as oversold or nearly oversold. Only in bear markets do these indicators remain at these levels for an extended period of time. Thus, as the chart of Selling Volume vs. Buying Volume and chart of performance of the Equal Weighted relative to weighted index also show, the Shanghai Composite index is at a critical juncture. Further weakness without a multi-week rebound in prices would be more indicative of a bear market than bull market.

If there is one caveat I would stress, it would be the possibility of a bear trap. There are so many investors monitoring this index that a brief break of support could lead to aggressive selling as an exhaustion move rather than a new trend. In that case, the index could decline to around 2500 and then reverse back above the 2640 support level and higher. We are not predicting a bear trap, but it is important for investors to recognize one when it occurs.

Disclosure: No positions