Asymmetric Investment Returns - Insights into portfolio management beliefs, strategies, and tactics that aim to create a positive imbalance between risk and reward

So says David Wilson at Bloomberg in his recent article: Extreme Pessimism Sets Stage for U.S. Stock Rally: Chart of the Day

He offers a nice chart showing two sentiment polls and explains:

The CHART OF THE DAY displays the results of weekly surveys by the National Association of Active Investment Managers and the American Association of Individual Investors since the beginning of 2009. Each appears in a separate panel.

This week’s reading for the manager index was 13.47, the lowest since March 2009, when the latest bear market in stocks ended. Survey responses can range between 200, indicating that managers are borrowing to profit from stock-market gains, and minus 200, showing the use of leverage to bet against shares.

The top panel tracks the active-manager readings, and the bottom panel shows the percentage of bulls among respondents to the survey of individual investors. The latter dropped this week to 20.9 percent, also the lowest level in 16 months.

Source: Extreme Pessimism Sets Stage for U.S. Stock Rally: Chart of the Day, Bloomberg.

Of course, investor sentiment is a topic I've mentioned a lot the past few months since it has swung from one extreme to another. Back on April 27th, before the -17% decline in market indexes, I posted Crowd Sentiment is Optimistic: No Surprise to See at Least a Short Term Price Decline. Investor optimism had hit an extreme optimistic level, and sure enough it preceded declining stock prices. As evidenced by sentiment polls like those in the chart, since stocks have now fallen the optimism changed to pessimism. We've also seen a lot of magazine covers with bearish tones, for example, the one I mentioned in Magazine Cover is a Bullish Signal for Crowd Sentiment.

As I've mentioned before, in order for prices to reverse back up to a new rising trend, buying demand simply needs to overcome selling pressure. These sentiment polls continue to show that investors, both professional and individuals, feel negative about the direction of the market. As their sentiment gets to an extreme they tend to over-react, eventually driving prices low enough to attract buying demand from those of us who raised cash when the crowd was extremely bullish.

However, there is no better measure of investor sentiment than the current direction of prices. As I've stated previously, sentiment alone isn't the best signal. The crowd can be right for some time. Sentiment can always get more extreme. Sentiment is confirming evidence of price action - don't fight the tape. The actual price direction is the judge and the jury, so that's our main guide. It appears in the days ahead we'll likely see if sentiment is negative enough to set the stage for a continuation to the upside, or if prices need to go lower to attract sustained buying interest.

The following chart is from my friend Ed Easterling at www.CrestmontResearch.com. I'm proud to say I was one of the very first to read and comment on Ed's book back in 2005: "Unexpected Returns". Prior to 2005, we had completed extensive research on historical Secular and Cyclical Bull and Bear Markets. Primarily, I had studied the trends going back over 100 years and tested tactical systems across those various trends. Our conclusion was that long term (Secular) Bull and Bear market's had occurred over and over throughout history. When Ed came out with his research in his book, we compared notes. His work was more focused on why Secular trends occur and under what circumstances. "Unexpected Returns" is a necessary study to understand the big picture. I can tell you that it was my understanding of market trends that motivated us to create the money management systems we benefited from the past five years. You can probably see how the timing of Secular situation correlates with our tactical asset management style - we built the Ark before the flood.

Though we studied and tested decision-making systems on much more detailed data, I thought I would show what these trends look like. Clearly, the U.S. stock market has been in a Secular Bear Market for a decade. Based on history, the current Secular Bear could last as long as another decade. Notice this chart goes back to 1900.

Below we show what the last Secular Bear Market looked like. This chart appeared in 5 of our quarterly portfolio commentaries before the waterfall in 2008. We continued to discuss these cycles so our investors were prepared for the possibilities. The last Secular Bear Market was about 16 years. We call the cycles oscillating up and down over time Cyclical Bull and Bear Markets. They typically last 1 - 4 years. At this point, the U.S. stock market has been in a defined Cyclical Bull Market since March 2009, however it is currently under pressure.

You can probably see the risk of a passive strategy and the potential value of actively managing risk and dynamically adapting to these systematic directional price trends. It doesn't require perfection.The average uptrend was 38%, the average decline was -25%. It was 343% cumulative peak to trough drifts, an average of 21% a year.

From this wisdom, we have some understanding of what is possible. Sadly, investors who do not actively control their risk may experience more waterfall losses and possibly panic selling before the current Secular Bear is over.Of course, just "being" tactical isn't enough. Portfolio management is a craft that requires wisdom, skill, and experience. Like any craft, it's learned over time.

"Those who cannot remember the past are condemned to repeat it"

- George Santayana

In a post titled Golden Cross Goes "Dark". Barry Ritholtz over at The Big Picture, points out the "Dark Cross" signal has occured on the NYSE Stock Index. It is sometimes called a "Death Cross". A Dark Cross signal means the 50-day moving average has crossed below the 200-day moving average on a major stock index. He quotes a comment from Mary Ann Bartel regarding the current status:
 

June 23, 2010 marked the 1-year anniversary of last June’s bullish Golden Cross of the 50-day moving average above the 200-day moving average. This Golden Cross signal preceded a 12-month return of 22.4% on the S&P 500. The average 12-month return for the 42 Golden Crosses that have occurred since 1928 is 9.6%. More importantly, the June 23, 2009 signal occurred during the NBER recession that began in December 2007 and Golden Crosses associated with recessions show a much stronger average 12-month return of 19.5%. The average 12-month return for the S&P 500 over the same period is 7.2%.[...]

The bearish counterpart of the Golden Cross is called a Dark Cross. This signal occurs when the 50-day moving average crosses below the 200-day moving average. For the S&P 500, Dark Crosses are not all that bearish. The 42 Dark Cross signals that have occurred since 1928 have generated an average 12-month return of 2.4% for the S&P 500 vs. the average S&P 12-month return of 7.2%.[...]

 

Below we show that the NYSE Composite Index has indeed signaled a Dark Cross.

^{Chart courtesy of eSignal}

However, we point out that the S&P 500 index hasn't yet crossed. We also point out that the current price is "at" the point we previously called "the line in the sand" we most recently discussed here. The current price range has been supported by buying interest at the February low and the more recent lows in May. So, I will suggest this index is at an important spot. If it does indeed move below the support line, it changes the short term trend to lower highs, and lower lows (a down trend). If the price continues to drift much lower, it could also signal a Dark Cross. Though the signal itself isn't a "cause" for futher systematic price decline, it would be evidence that suggests a reversal of the rising trend over the past year has ended, as defined by the Golden Cross (50 > 200) and then a Dark Cross (50 < 200). What does history tell us about these signals? Keep reading...

Our friends Ron Griess and Mark Cremonie over at www.thechartstore.com did some analysis on the S&P 500 Composite from 1930 and present the following two tables. The first table shows the performance of the S&P 500 Composite for the time periods listed when the 50-day moving average is falling and crosses the 200-day moving average while the 200-day moving average is still rising.

The second table shows the performance of the S&P Composite for the time periods listed when the 50 day moving average is falling and crosses the 200 day moving average and the 200 day moving average is also falling.

 Like many things in life, success has nothing to do with predicting the future. It's more about responding and adapting to things as they unfold. At this point, it seems there's a good chance this price level will hold since investor sentiment is rather negative (which is a positive). But if it doesn't, this will define a trend change. It's how we adapt to it that makes all the difference... 

We hear people talking a lot about stock indices like the S&P 500 stock index, referring to the index more than just a proxy for stocks, but also an investment. What you see below is a monthly chart of the past 10 years for the S&P 500 stock index. It is an index that many investment managers benchmark. I look at this chart, and I wonder; Who wants this?
 

Ten Year Chart of the S&P 500 Stock Index

^{Chart courtesy of eSignal}

In August 2000 the S&P 500 index was as high as 1,525. On Friday, it closed at 1,064. This index is down -43% from were it was ten years ago. (1065 - 1525) / 1064 = -43%. We could say the same about its more recent high point in October 2007.

At Shell Capital, we don't benchmark indexes. Our objective is to earn as much profit as we can with a specific amount of absolute risk. We call this "absolute return" rather than "relative return". We do not have benchmark-itis and we avoid relativity.

In a post back on April 29th, we shared a study we did on price reversals of 10% or more on the S&P 500 stock index. We titled it:10% Reversals in the S&P 500 Over The Past Decade. In this study, we showed that during the 2003 - 2008 Cyclical Bull Market the S&P 500 didn't decline more than -10%. But when it did, it was within the Cyclical Bear Market that occured in 2008 - 2009. This is a part of a much larger study our quantitative research has conducted. Since 1928, the S&P 500 stock index has declined more than 10% on 93 occasions. That's a little more than once a year, on average. Over that 82 year period, when this stock index declined 10%, its average decline was -19.57%, its median -16.39%. Of those 82 declines, 33 were -20% or more, which is commonly labeled a "bear market". That is, about 35% of the time these -10% corrections eventually became "bear markets". What does this mean? It is useful to study history so that we have some frame of reference about a range of possibilities. George Santayana (1952) said "Those who cannot remember the past, are condemned to repeat it." What do we learn from this? -10% or more declines in the stock market happen, and sometimes they get larger. We prefer to employ our active risk management systems to avoid the larger ones, those that are beyond our risk threshold, by decreasing exposure to the possibility of loss.

Since we shared that study, the stock market has reversed down into what is currently defined as a "correction". To update the data from that study, we share the below comparison of Major Market Indices since April 26th. As you can see, small company stocks as represented by the Russell 2000 index and the S&P 600 index have declined the most, so far.

Source: Stockcharts.com

Below we show a list of U.S. sectors.

Source: Stockcharts.com

U.S. commodity groups are also in short term declining trends, with the exception of Precious Metals.

 Source: Stockcharts.com

International stocks are also in corrections. Two world markets are in "bear market" territory (as commonly defined as a -20% decline). Those two markets are Austria and Italy with the Austrailia index coming in a close 3rd.

 Source: Stockcharts.com

And finally, we show currencies including the U.S. Dollar and selected foreign currencies around the world.

 Source: Stockcharts.com

So, it's safe to say that the world markets are at least in a short term declining trend. Of all these world markets, the only that are positive is the U.S. Dollar, the Japanese Yen, and Precious Metals. It is actually unusual for the U.S. Dollar and metals to be correlated (trending in the same direction) but not entirely unusual on a short-term basis. Clearly, active risk management to some degree has been useful since we commented near the price peak that Crowd Sentiment is Optimistic: No Surprise to See at Least a Short Term Price Decline (April 27th)  and Today’s Stock Market Decline is about Twice a “Normal” Move (April 27th).

We should note that these charts are not designed to signal tactical trading decisions, but instead to simply illustrate the recent returns (or lack of) for a range of various world markets.

As a follow up with yesterday's comment "Drawing the Line in the Sand for the Stock Market: A Look at the S&P 500 Index Trend "... so far, so good. As you can see in the chart below, the February lows held today. The stock indexes were down 2-3% most of the day. But by the day's end, the selling pressure dried up. The index closed very close to yesterdays closing price. For those who are looking for a reversal up, you "may" be in luck soon. I say this because of today's price action.

If you look at the blue arrows, those are called a "Hammer" on a candlestick chart. The Hammer is a common bullish reversal pattern that forms after a decline. It's common enough that a Hammer marked the February low. In addition to a potential trend reversal, hammers can mark bottoms as it did in February. The height of the candle shows the wide high to low price range that traded today as sellers drove prices lower during the session but buyers stepped in at the end of the day and took over. Hammers are similar to selling climaxes. We'll want to see confirmation, which could be a gap up in the days ahead. For prices to reverse back up, the selling pressure that drove it down simply needs to be overwhelmed by buying demand. For that to occur, prices have got to get to a low enough level to attract buying interest.

 Chart courtesy of eSignal Pro

Times like these can be very difficult for the passive investors who remain fully invested all the time. They are caught in a loss trap and don't know for sure how long it may last or how deep the losses may get. If the current price range does not hold support (the red highlighted area on the chart) and reverse up, the pressure will get stronger and stronger as their losses mount. There is also a possibility that stocks could reverse up for a few days or weeks and then reverse down to eventually break the February lows. From that point, it could keep going to the March 2009. Anything is possible. This is why, at Shell Capital Management we actively manage our risk by increasing and decreasing exposure to the possibility of loss... When prices are peaking out or start reversing down and volatility is rising, we exit weak positions systematically to reduce our exposure to loss. The market isn't the risk. The risk is the exposure. If we have no exposure, we have no risk. Our risk control system is a huge advantage for us. It allows us to control our risk to a maximum amount we're willing to take as we aim for capital gains. You see, we can't fully hedge or eliminate our risk all the time and expect to earn a profit. We aren't unwilling to take any risk at all. We must employ "some" risk in order to have profit potential. In order to earn a profit, we need exposure to some degree at some point. What we do differently is control the "degree of exposure" and the "point". We only want exposure when the odds are overwhelmingly in our favor. We do it by buying and selling things...

As it turns out, we have something in common with many of the most famous investors in the world. The following statement was written by Warren Buffett in a 2003 "Letter to the Shareholders of Berkshire Hathaway".

 

When we can’t find anything exciting in which to invest, our “default” position is U.S. Treasuries, ... Charlie and I detest taking even small risks unless we feel we are being adequately compensated for doing so. About as far as we will go down that path is to occasionally eat cottage cheese a day after the expiration date on the carton.*

It appears that Mr. Buffett, like us, has an objective of absolute returns ...

We like avoiding large losses on our portfolio (as defined by -15% or more). We like not having to "worry" if this Hammer is indeed a setup for a reversal, or not. We reduce our exposure to loss to an acceptable level and then go with the flow...

 *Source: http://www.berkshirehathaway.com/letters/2003ltr.pdf

Yes, you read that correctly. When investors get scared, that's a good thing for stocks. When investors are optimistic, that's negative. That's just the way it works. And we're going to show you here, over and over again.

A few weeks ago in Crowd Sentiment is Optimistic: No Surprise to See at Least a Short Term Price Decline I noted that our measures of investor sentiment signaled a high level of optimism that historically precedes a correction for the stock market. When “the market” gets too excited, as they often do after prices have risen in the recent past, that’s about the time a rising trend in stock prices reverses to the downside. At Shell Capital, we like to go with the flow of the prevailing trend by staying on the right side of the directional drift and then we adapt to directional changes in price. That is, we don’t fight the tape. But when we see signs of people getting to greedy, that’s when we start to “set up” and pay attention- aware of a potential reversal. Though we have quantitative systems that measure a wide range of investors’ fear and greed, we also get important signals from the comments from people. For example, we pay attention when comments from investors are about how much money they’ve made or could have made, or should have made in the recent past. When optimism gets to an extreme, we aren't surprised when prices move in the opposite direction. And when the direction does change, we're changing with it. A few weeks ago, I suggested that investors' high level of optimism suggested a correction was likely.

But that was then, this is now. A fundamental edge we have here at Shell Capital is that we live in the "now". We make decisions in moments of "now". We can do nothing in the past. We can only do things now, or not now: that’s the only choice we have. A quarterback can only throw the ball right then. He can’t do it the day after the game. He can’t do it in the future, either. He can visualize how he’ll throw the ball in the future, and that may be useful mental rehearsal for him and mental rehearsal may be a part of his advantage. But when the time comes, he can either throw, or not throw. And the choice is right now. We live in moments of now. It's a big advantage for me: I know the secret to human decision-making: it’s now, or not. That's the choice. You can probably see how simple my decisions are...

In the table below we show the S&P 500 Stock Index when it reversed down -10% or more over the past 10 years. Declines of -10% or more are not very common: they primarily happen during Cyclical Bear Markets. Out of 11 such declines, ALL OF THEM occured within a Cyclical Bear Market or marked the beginning of one. Over the past decade, when the stock index does decline -10%, on average it declined -22.3%. (Note: A -20% decline is commonly called a "bear market"). As a side note in regard to the average: 11 events isn't enough to be statistically significant, so we look back 82 years. Since 5/14/1928 there were 92 declines of more than -10%. On average, they lasted 102 days and the average decline was -19.65%. That isn't a lot different than the last decades mean of  -22.3% over an average of 100 days. You can probably see how we use studies like these to quantify, define, and understand, what is normal, or not.

10% Gains also occured 11 times over the past decade. IN ALL INSTANCES, the 10% or more upward drifts occured within a bear market or it marked the end of a Cyclical Bear Market and the beginning of a new Cyclical Bull Market (no one knew that then, so we could also state that all instances occured within a Cyclical Bear Market). For example, the first low point that made the list was on 4/14/2000. The 3/24/2000 close of 1527 had previously marked the highest point and the beginning of a Cyclical Bear Market that lasted through the lowest close on 3/11/2003. That same date marked the beginning of the Cyclical Bull Market the continued through it's end on 10/9/2007. You may notice that these dates are also marked on the table below since they were "picked up" by the 10% swings. Of course, the end of a Cyclical Bull Market is the beginning of a Cyclical Bear Market marked by that same date. Finally, the last 10%+ gain in the index started on 3/9/2009, which we now know was the beginning of the current Cyclical Bull Market as the Primary Trend today.

When the index gained 10% which it did 11 times over the past decade, its average gain was 30.52%. That isn't far from the 33.7% average gain that occured the 93 times the index gained 10% in the past 82 years going back to 2/20/1928. And, during the past decade the move averaged 234 days and that is comparable to 222 days over the past 82 years. We could suggest that this data is more meaningful when we separate the it by cyclical trends. For example, we could remove the Cyclical Bull Markets (3/11/2003 - 10-9/2007 and 3/9/2009 -NOW) and probably have more meaningful information.  Without those two cycles, the 10%+ move averaged 55 days and the average gain was 18%. You can probably see how far we could take this to better quantify, define, and understand, directional trends... We'll build on this, later.

Over the past decade, the S&P 500 Stock Index has lost -17.97%. That's an annualized loss of -1.96% over the past 10 years. CLEARLY, a passive buy and hold strategy hasn't been profitable. Below we show the 10 year trailing return (or lack thereof) for this popular index and its 4 different formats. The most commonly quoted capitalization weighted index, the Total Return (includes dividends), the Equal-Weighted (all 500 stocks are weighted equally, rather than based on size), and the Equal-Weighted Total Return. None of them has achieved a positive return of the past decade.

S&P 500 Stock Index (4/28/2000 - 4/28/2010)

S&P 500 Total Return (4/28/2000 - 4/28/2010)

S&P 500 Equal-Weighted Index (4/28/2000 - 4/28/2010)

S&P 500 Equal-Weighted Total Return Index (4/28/2000 - 4/27/2010)

Relative Return managers,  money managers whose objective is to track and outperform an index like the S&P 500 often make statements like "Time - not timing - is on your side" or “It’s not timing the market, but time in the market.” They use a one-sided study showing that missing the best days (the largest daily gains) of the stock index leads to poor returns. While that is obviously true, that is a misuse of data and it's only one side of the story. First, we could probably win the lottery twice in a year easier that we could miss the exact 10 largest gaining or losing days in the stock market on an annual basis. Second, what they don't tell you is that missing the downside creates the kind of positive imbalance between profits and losses that we want. Since we're talking about the past decade, missing the 10 worst down days was the only way to achieve a positive return... and the more downside we missed, the more positive the outcome.

This is just one of many examples that we will share that explains the logic behind why we employ our active risk management systems. I will suggest that not managing risk would be illogical. Isn't it fascinating that the very people who presume that all investment decisions are made with logic are the same people who believe that markets are efficient? So they believe all they need to do is buy and hold a index like the S&P 500. And, relative return managers whose objective is to try to beat a benchmark index don't manage absolute risk,either. Instead, their idea of risk isn't the loss of money at all. For a relative return manager, risk means "tracking error". For example, a manager whose benchmark is the S&P 500 could hold all the 500 stocks in the index at the exact same weight of the index and would consider that no risk. That is, as long as they hold the stocks in the index, they will track the benchmark and have no risk of tracking error. Stock indexes are fully invested at all times. That necessarily means an index is fully exposed to the possibility of loss with all the capital, all the time. Therefore, a Relative Return manager whose objective is relative performance vs. a benchmark stock index would actually consider it a risk if he or she held 10 or 20% of the portfolio in cash. That is because their risk is not tracking the benchmark. It has nothing to do with losing money. When these managers don't track close to their benchmark, they call it 'tracking error risk" or "style drift". Therefore, when the indexes' decline a lot, so will they.

I'm just pointing out a few "illogical inconsistencies". If you've ever wondered why the vast majority of investment managers closely track the daily moves of the stock indexes, now you know. This is in sharp contrast to our strategy at Shell Capital. Instead of a market-based strategy that tracks the moves of a stock market index, our first priority is capital preservation. Because I can control my risk, that allows me to focus on the momentum in our profits. We believe that logical investors are risk averse. Investors perceive downside volatility worse than upside volatility. We believe that is logical and wise because of the exponential nature of losses. Now, risk control shouldn't be confused with the elimination of all risk. If we neutralize all risk there would be no reward. We must take "some" risk in attempt at profits. The difference in our method is that we know what our risk is, and we control it, so our focus is on creating a risk/reward profile that meets our investors' objectives. Clearly, that cannot be accomplished with a passive strategy or a relative return strategy...

In the table below we show the S&P 500 Stock Index when it reversed down -10% or more over the past 10 years. Declines of -10% or more are not very common: they primarily happen during Cyclical Bear Markets. Out of 11 such declines, ALL OF THEM occured within a Cyclical Bear Market or marked the beginning of one. Over the past decade, when the stock index does decline -10%, on average it declined -22.3%. (Note: A -20% decline is commonly called a "bear market"). As a side note in regard to the average: 11 events isn't enough to be statistically significant, so we look back 82 years. Since 5/14/1928 there were 92 declines of more than -10%. On average, they lasted 102 days and the average decline was -19.65%. That isn't a lot different than the last decades mean of -22.3% over an average of 100 days. You can probably see how we use studies like these to quantify, define, and understand, what is normal, or not. 

10% Gains also occured 11 times over the past decade. IN ALL INSTANCES, the 10% or more upward drifts occured within a bear market or it marked the end of a Cyclical Bear Market and the beginning of a new Cyclical Bull Market (no one knew that then, so we could also state that all instances occured within a Cyclical Bear Market). For example, the first low point that made the list was on 4/14/2000. The 3/24/2000 close of 1527 had previously marked the highest point and the beginning of a Cyclical Bear Market that lasted through the lowest close on 3/11/2003. That same date marked the beginning of the Cyclical Bull Market the continued through it's end on 10/9/2007. You may notice that these dates are also marked on the table below since they were "picked up" by the 10% swings. Of course, the end of a Cyclical Bull Market is the beginning of a Cyclical Bear Market marked by that same date. Finally, the last 10%+ gain in the index started on 3/9/2009, which we now know was the beginning of the current Cyclical Bull Market as the Primary Trend today.

When the index gained 10% which it did 11 times over the past decade, its average gain was 30.52%. That isn't far from the 33.7% average gain that occured the 93 times the index gained 10% in the past 82 years going back to 2/20/1928. And, during the past decade the move averaged 234 days and that is comparable to 222 days over the past 82 years. We could suggest that this data is more meaningful when we separate the it by cyclical trends. For example, we could remove the Cyclical Bull Markets (3/11/2003 - 10-9/2007 and 3/9/2009 -NOW) and probably have more meaningful information. Without those two cycles, the 10%+ move averaged 55 days and the average gain was 18%. You can probably see how far we could take this to better quantify, define, and understand, directional trends... We'll build on this, later.