Will Small Cap Out-performance Continue?
(January 29, 2006)
Dear Subscribers and Readers,
Here is an interesting fact: It is generally well-known that if Wal-Mart was a sovereign country, it would be the 8th largest trading partner with China. Given this fact, there is a misconception that aside from groceries, virtually everything that Wal-Mart sells is imported from China – and thus any further revaluation of the Chinese currency (the Renminbi) would significant hurt Wal-Mart's bottom line. Nothing could be further from the truth. As a matter of fact, Wal-Mart only imported (directly and indirectly) $15 billion worth of goods from China in 2004 – which amounts to less than 15% of what it buys from its suppliers.
We switched from a 25% short position in our DJIA Timing System on the morning of October 21st at DJIA 10,265 – giving us a gain of 351 points from our DJIA short on July 14th. On a 25% basis, this equates to a gain of 87.75 points. We switched to a 25% short position in our DJIA Timing System shortly after noon on Wednesday at DJIA 10,840. We then switched to a 50% short position (our maximum allowable short position in order to control for volatility in our DJIA Timing System) on Thursday afternoon at DJIA 10,900 – thus giving us an average entry of DJIA 10,870. As of the close on Friday (10,907.21), our position is 37.21 points in the red – but at this point, we believe that the correction in the stock market is not over yet.
The big news last week was the shocking preliminary GDP report last Friday – which reported a staggeringly low 1.1% real GDP growth for the fourth quarter of 2005. Most of this surprisingly slow growth was blamed on a huge drop in durable good sales (especially auto sales), a tepid growth in business investments, as well as “one-time related” special events such as an increase in oil imports due to Hurricane Katrina and Rita. The consensus reaction among economists is that this was such a temporary blip (some went further and stated that the GDP number will be revised back to the 3% level in the next release) – and that GDP growth will recover back to the 3% to 4% level for the entire 2006 year.
Given our mid-cycle slowdown scenario, this author is skeptical. I have previously gone over our reasons for this skepticism – but let's briefly recap these reasons. First off is declining global liquidity – given that the ECB should hike its ST rates two to three times for 2006 and given that the BoJ should also start raising rates sometime this year. Second off is the deflation of the U.S. housing bubble. Given the subsequent decline in mortgage equity withdrawal (MEW), consumer spending should be expected to decline 1.5% to 2% in the upcoming 12 months. Please keep in mind that this estimates ignores “multiplier effects” – such as the loss of real estate related jobs due to a softening market as well as the negative “wealth effects” caused by a slowing housing market. A third reason involves some circular logic: Unless the U.S. and China economies slow down significantly starting in the next few months, there is a good chance that commodity prices will continue to rise – thus further squeezing corporate profits and decreasing consumer spending going forward. In other words, this author believes that there is “no way out” of this mid-cycle slowdown scenario – although I am not yet willing (unlike George Soros) to declare that the U.S. will experience a recession in 2007. Globalization and “offshoring” can only take you so far. Many Americans like to believe that labor in both China and India are limitless and interchangeable. Logic will dictate that this is not the case. Given that corporate profits as a percentage of GDP in the third quarter of 2005 was the highest since 1997 (and prior to that, 1969), my question is: How much further can cost-cutting go? In fact – while I have discussed the virtues and continuing trend of globalization in many of our past commentaries – there is a good chance that U.S. labor costs will start to enjoy a sustainable rise sometime this year. The whipping of the pilots' and auto workers' unions may have marked the high point of the power of capitalists and shareholders for this cycle.
Let's now get on with our commentary. Anyone that has read a book on stock market history should know that small caps have historically outperformed large caps in the U.S. stock market. This “size effect” has also been observed in many international markets – even when adjusted for risk. That being said, subscribers should know that such out-performance by the small caps is not consistent every year. In fact, history has shown that relative out-performance or under-performance of small caps relative to large caps have been relatively cyclical. That is, just like the stock market or bonds, over-performance of small caps relative to large caps also tend to experience sustained bull and bear markets.
Before I go into further details, please keep in mind that all of the following data are courtesy of Ibbotson Associates. Their datasets (which are updated once a year and published in the “SBBI Yearbook”) contain the most comprehensive history of the U.S. domestic stock market – and its reliability is second to none. While the SBBI Yearbook does contain large company stocks data prior to 1926, there does not seem to be any reliable data for small companies prior to that year. Therefore – for the purpose of this commentary – let's begin our observation of small caps vs. large caps starting from the year 1926 and ending in 2004. Note that the 2006 SBBI Yearbook will be released later this March – and once this author gets his hands on a copy, I will be sure to update this study for our readers as soon as possible.
Without further ado, following is the annual chart showing the cumulative returns (for $1 invested, and with dividends reinvested) of large caps and small caps from 1926 to 2004, courtesy of Ibbotson Associates:
Looking at the above chart, there are two immediate implications:
- First of all, small caps have outperformed large caps by a greater than 5-to-1 ratio since 1926 – and have performed especially well relative to large caps over the last 40 years. This “size effect” is significant, in that not all of the excess returns of small caps over large caps can be simply explained away by the fact that investors are incurring more risk by buying shares of small cap companies.
- The second implication is much less obvious – and may be difficult for the uninitiated to observe this on the above chart. From 1926 and up until 1957, small caps have actually underperformed large caps on a relatively consistent basis! Why has this been the case? Throughout the 1920s and up until the Fall of 1929, many U.S. investors were willing to buy shares of small and mid-sized companies believing that they can perform or sustain themselves as well as large-sized companies during economic recessions. The huge bear market and subsequent Depression of September 1929 to July 1932 totally dispelled that belief, and for the next 30 years, investors simply refused to buy shares of small cap companies unless they were trading at significant discounts to shares of large cap companies. Moreover, more stringent reporting requirements as required by the SEC after the 1929 to 1932 debacle have also done much to improve the quality of smaller companies listed on the NYSE – thus enabling better potential returns in shares of small companies going forward.
Going forward, it may be reasonable to believe that an “intelligent investor” can simply put all his money in small cap index funds and then simply forget about them until the day he retires. But such a strategy may not be such a wise one – at least for the next few years – as small cap out-performance has also been mired in bull and bear markets ever since 1926. For example, large caps outperformed small caps in every year from 1926 to 1931. They will go on to beat small caps again in every year from 1946 to 1948, and again (with the exception of 1954) from 1951 to 1957. Just like the fact that stocks have outperformed bonds on a long-term basis but underperformed for significant periods of time, it is definitely not a given that small caps will out-perform large caps during each and every year. Following is an illustration of historical periods where small caps have outperformed large caps on a consistent basis. Note that the prerequisite is a period of out-performance of at least two consecutive years:
While final statistics have still not been published by Ibbotson Associates, it is interesting to note that Russell Investment Group (keeper of the Russell stock market indices) have already issued a statement claiming that large caps did indeed outperform small caps during 2005. If this is indeed the case, then the 1999 to 2004 period of small caps out-performance would have officially ended.
However – and more importantly – subscribers should note that small cap out-performance tend to end when one of two things (or both) occurs:
- Small cap out-performance tends to end after a sustained period where small caps outperformed large caps. Okay, this is rather obvious – but please note that from 1999 to 2004, the cumulative returns of small caps beat that of large caps by 135%. While this percentage is just slightly above the average of those during the last 70 years, it is definitely nothing to scoff at, and given that the duration of small cap out-performance is also slightly above the average of 5.13 years, small cap investors should be very careful here.
- Small caps also tend to significantly under-perform large caps during periods of liquidity squeezes or in the face of external economic shocks. Such memorable years include 1929, 1930, 1937, 1973, 1990, and 1998. This is not surprising – as the businesses and balance sheets of smaller companies tend to fare worse during liquidity squeezes.
Given the fact that small caps have out-performed large caps for a total period of six consecutive years from 1999 to 2004, and given that global liquidity is declining as we speak, this author is not exactly thrilled with investing in small cap funds over the next few years. Moreover, it should be noted here that small cap sentiment among retail investors is also extremely bullish – given that mutual fund inflows to small cap funds have been extraordinary over the last few years. In fact, not only have investors been pouring their cash into small cap funds, but they have also been liquidating their large cap holdings in order to purchase small cap funds!
In terms of valuation, small caps in general are not so hot either. A press release from JP Morgan during the third quarter of 2001 stated the following about small caps: “In fact, small cap stock valuations are near a 20-year low. At the peak of the last great bull market in small cap stocks (1975-1983), small caps, as measured by the median PE ratio of the Russell 3000 Index, sold at a 60% premium to large caps, as measured by the weighted average PE ratio of the Russell 3000. Today, the weighted average PE ratio of the Russell 3000 is 28.3X, while the median PE ratio is 14.8X. This ratio represents one of the greatest small cap discounts since the index's creation, and is even more dramatic than in 1973, just prior to the last small cap boom.” While this is not a straight comparison, it should be noted that since the third quarter of 2001, the P/E ratios of small caps in general have risen significantly relative to large caps. As measured by the 12-month trailing P/E ratio of the S&P 600, small caps are now approximately 20% more expensive than large caps (S&P 500). In the author's opinion, small caps should, in general, significantly under-perform large caps over the next few years.
That being said, subscribers who are good at stock selection should not be deterred from investing in small caps going forward – since, in the case of investing in small caps, stock selection is really the key. Precisely because of this, subscribers who are investing in individual small caps should at least have 10 to 15 stocks (and in different industries) in his portfolio in order to properly diversify his portfolio.
Looking at the stock market from a shorter-term timeframe, this author still does not currently like the stock market, as it is still very overbought as measured by many of our indicators, such as the equity P/C ratio, the NYSE McClellan Summation Index, and so forth. While this author recognizes the fact that company buybacks and private equity takeovers have added much liquidity to the stock market over the last few years, readers should note that global liquidity is declining – and thus the era of cheap money is going to end sooner or later. Case in point: Last year, there were nearly $500 billion in global private equity deals, and yet the U.S. stock market did very little. Can we expect the same amount of activity this year, given that CALPERS have publicly declared that private equity is in a bubble and that they have cut back on their private equity holdings back in October 2005? Short of anything that resembles another record, the U.S. and global stock markets should disappoint this year. Moreover, it is interesting to note that hedge funds collectively experienced outflows during the fourth quarter of 2005 – the first time that hedge funds have actually had more redemptions than investments in over a decade. Is this another sign of deteriorating global liquidity? Time will tell.
Moreover, cash levels at mutual funds only stood at 4.1% at the end of November 2005 – and chances are that we will be at or below the 4.0% level again at the end of this month. February is also a month where we traditionally see heavy insider selling in the U.S. stock market – and this year should be no different given the many disappointing earnings and 2006 forecasts for many of the U.S. large caps. While earnings for small and mid caps have turned out relatively fine, it should be noted that their 2006 earnings forecasts have been less than stellar.
Going back to our discussions on consumer spending and our mid-cycle slowdown scenario for 2006, I now want to take a little bit of time and discuss the relative strength of the Retail HOLDRS (RTH) vs. the S&P 500. Readers who have kept track of our commentaries should know the emphasis we have placed on the relative strength of the RTH vs. the S&P 500 as a leading indicator of the broad market. Ever since this cyclical bull market began in October 2002, the relative strength of the RTH has typically led the broad market anywhere from two to eight weeks. Following is the latest weekly chart showing the relative strength of the RTH vs. the S&P 500:
As can be seen from the above chart, the latest weekly reading of the relative strength of the RTH again slumped in the latest week. Ignoring the reading from two weeks ago, this latest reading represents the lowest reading since March 2003 – carrying significantly bearish implications for both the broad market and consumer spending in the weeks ahead. Whether this reading will hit a new low in the upcoming week will depend a lot on the reaction to Wal-Mart's latest January sales growth report. While this latest report looks relatively bright, keep in mind that most of the gains came from food sales as opposed to general merchandise – and the former tends to carry a significantly smaller profit margin than the latter. Also, as I am writing this, crude oil is again trading at above $68 a barrel, and natural gas at near $9/MMBtu. Should energy prices continue to power ahead in the next few weeks, there is no doubt that the relative strength of the RTH vs. the S&P 500 will again hit a new low.
Let's now end with a discussion of the most recent action in the stock market. In last weekend's commentary, I had stated: “While nearly everyone is focused on the huge 292-point decline in the Dow Industrials and 26-point decline in the S&P 500, it is important to note here that the Dow Transports actually rose 11.38 points in the latest week – which qualifies as one of the most flagrant short-term divergence in Dow Theory history. In other words, the Dow Transports is not confirming the Dow Industrials on the downside. This is doubly significant given that the Dow Transports has been one of the most reliable leading indicators of the major market indices since this cyclical bull market began in October 2002.” I also expressed the view that if the Dow Transports was to hold above its January 17th closing low, “the bulls should be safe for now.” In retrospect, the relative strength of the Dow Transports foretold the rally over the last week. In fact, the Dow Transports actually made another new all-time high at the close last Friday – forcing us again to readjust our scales in the following chart. Following is the daily chart showing the most recent action of the Dow Industrials vs. the Dow Transports:
From a Dow Theory point of view, it should be noted here that the Dow Industrials still refuses to confirm the Dow Transports on the upside. In fact, the Dow Industrials has been underperforming the Dow Transports since two years ago. This is not a surprise, since we have been in a small and mid cap equity bull market since 1999 – but even in a small cap equity bull market, the large caps (and the Dow Industrials) should also be confirming on the upside by making all-time highs. Moreover, it is interesting to note that the Dow Utilities made an all-time high in late October 2005 but have been struggling ever since. Coupled with the fact that long-term bonds have been taking a beating over the last couple of weeks, this author believes that the stock market is now very vulnerable to a significant decline going forward. Again, the real test of the markets will come after the official appointment of Ben Bernanke as the new Chairman of the Federal Reserve on January 31st, as well as Google's earnings on January 31st and Starbucks' earnings on February 1st. This author's prediction: Google will disappoint no matter what earnings or guidance they give. Historically, large cap growth stocks have nearly always under-performed – and it should be no different with Google, especially given that earnings estimates have been consistently revised upwards over the last few months and are now sitting at unprecedentedly high, unreachable levels. Unless Google is added to the S&P 500, my guess is that Google will continue to underperform the general market over the next few months.
Let's now discuss our most popular sentiment indicators and start with the American Association of Individual Investors (AAII) Survey. In brief, while last week's AAII reading did get to an oversold area, it is essential to keep in mind that our longer-term readings (the ten-week moving averages) are still calling for a resumption of the recent correction sooner rather than later. During the latest week, the Bulls-Bears% Differential in the AAII survey declined from a 24% reading to a negative 2% reading – thus putting us in oversold territory on a weekly basis. Thus, the rally in the latter part of last week should not come as a total surprise, but given that the market did not really get oversold on the Friday before last (as indicated by our longer-term technical and sentiment indicators), my guess is that this current rally will be very short-lived. Meanwhile, the ten-week moving average decreased from an overbought reading of 23.3% to 19.9% - suggesting that this survey (and thus, the stock market) is now in the midst of embarking on a new downtrend. Following is the weekly chart showing the Bulls-Bears% Differential in the AAII Survey vs. the Dow Industrials from January 2003 to the present:
At this point, our longer-term indicator (the 10-week moving average) is still signaling for us to stay short – so for now, we will remain 50% short in our DJIA Timing System until the AAII survey “sells off” to an oversold area (which will most probably involve at least two weekly negative readings and a ten-week moving average below 10%).
At the same time, the Bulls-Bears% Differential in the Investors Intelligence Survey has officially turned down, as the ten-week moving decreased from 34.2% to 34.0% in the latest week. This reversal is doubly authoritative given that it is turning down from such an overbought level. Meanwhile, the latest weekly reading declined from 34.4% to 28.4%. Following is the weekly chart showing the Bulls-Bears% Differential in the Investors Intelligence Survey vs. the Dow Industrials:
Coupled with the fact that the trend of the AAII survey has also turned down, there is a very good chance that the readings in the Investors Intelligence Survey will not reverse to the upside again until we start getting very oversold readings again. For now, we remain comfortable with the 50% short position in our DJIA Timing System – especially given that the Dow Industrials has been one of the weakest indices since this cyclical bull market began in October 2002.
Meanwhile, the Market Vane's Bullish Consensus – after coming off of two consecutive 70%+ readings – declined to a more manageable but still-overbought 68% in the latest week. Meanwhile, the ten-week moving average rose from 69.3% to 69.7% - the most overbought reading since August 1997. A reading of 68% or below in the upcoming week would cause the ten-week moving average to reverse on the downside – thus officially giving us a “sell signal” on all three of our sentiment indicators. At this point, this author believes that the market is still in the midst of topping. Following is the weekly chart showing the Market Vane's Bullish Consensus vs. the Dow Industrials:
Given that all three of our popular sentiment indicators are now very or extremely overbought, and given that the ten-week moving averages of the AAII survey and the Investors Intelligence survey have reversed – we will remain 50% short in our DJIA Timing System until the readings of these surveys decline back to at least a moderately oversold level again. At this point, the author still believes the market has yet to finish its downside correction and thus is very vulnerable to a multi-week decline.
Conclusion: Given that small caps have significantly out-performed large caps since 1999 (135% from 1999 to 2004); given declining global liquidity, and given that retail investors' sentiment on small caps are now at an extremely bullish level, it is reasonably logical to believe that small caps will start to significantly under-perform large caps over the next few years. Historically, small cap out-performance – just like stocks in general – have gone through phases of bull and bear markets, and this time should prove no different. This author believes that the next buying opportunity will be in the large cap, brand name stocks – which should occur sometime later this year or during 2007.
In the more immediate term, readers should continue to watch the relative strength of the retail HOLDRS (RTH) vs. the S&P 500, as well as the outcome of the Fed meeting and Google's earnings on January 31st, as well as Starbucks' earnings on February 1st. For now, our most objective longer-term technical and sentiment indicators are still telling us that the market is in the midst of forming a significant top – and so we will remain 50% short in our DJIA Timing System until we start getting more oversold readings in our technical and sentiment indicators.
Henry K. To, CFA