1/16/2019

Performance Review - 2018

Performance of the Focused 15 Investing portfolios for 2018 was negatively affected by the rapid up and down moves in stock market prices in November and December.  By the end of the calendar year, most of the model portfolios posted losses.  The immediate cause of the rapid price changes late in the year seemed related to global trade concerns, mid-term elections, government shutdown, slowing global growth, and general Washington chaos. 

These oscillating concerns occurred at the end of the year against the backdrop of:
  • Unsustainably high stock market returns beginning in November of 2017, coinciding with investor optimism about US corporate tax cuts. 
  • The perception that the stock market was “old.” Many observers perceived that the bull market that began in 2009 was getting old and it was time for declines. 
  • A US stock market that had already displayed signs of peaking. The US stock market started sending warning signs about peak earnings (lower Price-to-Earnings ratios and persistently high Price-to-Book ratios) beginning in January of 2018.
CPM’s Market Resilience Index (MRI) conditions provide additional perspective.  During 2018, the measure that indicates the long-term trend of stock prices shifted from being decidedly positive to being decidedly negative.  This shift took place in May 2018.  From that time on, it was unlikely that the market would attain new highs before visiting dramatic new lows. 

From the MRI perspective, the age of the bull stock market was not 2018’s most distinct feature.  The bull market that began in 2009 was rejuvenated in 2015/6 with a moderate decline in what I have called a phantom bear market.  Therefore, the age of the bull was not a primary issue.

Instead, the degree of euphoria in late 2017 and in January 2018 was a distinctive feature.  The Macro and Exceptional Macro indicate levels of euphoria.  The level of the Macro MRI in January of 2018 was among the five highest over the last 100 years.  The other highly euphoric weekly readings were 8/21/1997, 9/28/1986, 10/21/1955, 9/13/1929.  This comparison is cautionary.  Large declines followed the periods of high euphoria in the cases of 1997 (the 1998 declines associated with LTCM crisis mentioned below), 1986 (the large 1987 decline), 1929 (the stock market crash of 1929).  Regarding the Macro peak in 1955, the market declined almost 20% in 1957.  While these are the most euphoric periods, there have been many highly euphoric over the last 100 years. 

Since we expect that a major price decline may follow a peak in the Macro MRI, it is reasonable to wonder if it would be best to sell stocks as the Macro MRI trends higher – before it peaks.  Testing over the last 100 years of market history indicates that selling in the presence of high euphoria is not a strategy that delivers high return and low risk.  Doing so causes one to miss the periods when high euphoria persists for long periods of time.  

Instead, it is best, in general, to sell after a peak in the Macro MRI has occurred.  For example, if we had sold US stocks at a level of euphoria (i.e., a level of the Macro MRI) that represented the highest point in the 2007-8 period, we would have missed most of the returns for 2017.  Diamond returned about 39% in 2017, and this level of return can compensate for the challenges of 2018. 

Our portfolios did well in 2017 because the algorithms were trained on the similar earlier periods, such as 1920–1929, 1945-1966, 1990-2000.  The experience over these periods indicate that it is better to pick up the pieces in years like 2018 than to miss the years like 2017. 

Timeline
  • We were invested in the stock market through 2017 and into early 2018. 
  • After the peak in the Macro MRI in early May, the Focused 15 Investing portfolios became more completely defensive in early July of 2018 after the Micro MRI had peaked. A “Harvest 2” designation ended at that time and the target weights called for zero weight in stocks in model portfolios based on the D5 signal set (i.e., signals driving Diamond and Zircon). The expectation was for declines in stock prices. 
  • On July 20, 2018, the algorithms recognized that the period of very low resilience passed without major declines. The algorithms started recommending higher target weights for stocks and prices did indeed move higher. My commentary reflected the deteriorating longer-term market conditions. 
  • By October 12, 2018, the algorithms indicated that the short-term period of resilience was nearing an end and issued the last “Harvest 2” week, which had indicated a good time to remove assets from the stock market. 
  • There was a bounce after the midterm elections (November 6) that occurred as generally expected. 
  • By November 23, the Diamond model portfolio had returned about lost about 1% year-to-date and was about 1% behind the performance of the default portfolio. 

To this point in 2018, the algorithms were performing in a manner that I consider to be within the range of normal behavior.
  • By November 30, the Diamond model portfolio had a return of about 2% for 2018. However, this was about 5% behind the return of the default portfolio. 
  • Through the rest of 2018 and in the beginning of 2019, we had false starts on a short-term rally in stock prices. 
The time to override the algorithms would have been right after the election. Our portfolios were defensive at that time for the right reasons. But the algorithms responded to the post-election bounce as the bottom of a short-term (Micro) cycle, which was not the case. Trade concerns and other news-of-the-day issues seemed to depress stock prices beyond responses typical of the last 100 years.

I reviewed the algorithms and their performance for 2018 and have these observations:
  • High Volatility (or Variability) Alone was Not the Problem - I evaluated the level of variability of stock prices in November and December. Some market observers have said that the level of variability is at historic highs. I have not found strong evidence to support this view. I evaluated weekly variability, daily variability, and changes in weekly and in daily variability. While high, the levels of variability were not unusually extreme. 
  • Reversals Without Establishing New Trends - Several price reversals in November and December did not initiate a new price trend as quickly as has been typical over the last 100 years. News-of-the-day elements seemed to have a big impact but conditions that developed since late 2017 are set the stage. Without these conditions, I believe the same November and December events would not have produced the same market dynamics. 
  • "Slow-Twitch" was Better for 2018 - Some of the Focused 15 Investing model portfolios avoided losses and did quite a bit better than their default mixes during 2018. The prime example of this is sg129, “Institutional: D5, LC/SC, Multi-sector Bonds”. The better performing model portfolios relied more heavily on algorithms that moved more slowly. Because of this, they stayed out of the market and were not affected by the exogenous forces in November and December. Consistent with the nature of those algorithms, those portfolios have low allocations to the stock market as of January 14, 2019, despite the rally in US stocks. These slower responding model portfolios have lower return than the others when viewed over a longer time horizon, as shown in the figure above. However, for the 2018, their level of responsiveness fit the events of the year nicely. 
  • None of the algorithms should be changed at this time - In general, I do not believe any of the model portfolios should be changed. The rotational portfolios based on the Macro Rotation Signal, which I describe in a blog post dated December 30, 2018, would not completely address the performance issues of 2018. Over the last several weeks, I have evaluated numerous other ways of using the Macro Rotation Signal. While there are some promising options, thus far, none are clearly superior to Diamond in terms of risk, return, and ease of implementation. I will provide updates on this research as it progresses. 
  • Losses Will Occur from Time to Time – The following section looks at the longer-term historical simulation of the current algorithms as used in Diamond. This information shows that there are periods of loss in otherwise attractive performance histories.
The Historical Perspective

The CPM investment strategies have been evaluated for their strength over 100 year of stock market history.  Over that time, there have been numerous periods of performance declines like what we experienced in 2018.  A recent one is 1998.  During the 1990s and especially in the late 1990s, US stock market gain were high.  During 1998, the surprise collapse of the Long-Term Capital Management (LTCM) investment manager rippled through the capital markets and required intervention from regulators and other financial institutions.  Performance of the model portfolios was poor during that period.  This is indicated by arrow “A” below.  Arrow “B” indicates the end of 2018. The blue line is the model portfolio.  The tan line is the default mix.


On the figure above, we can see that the model portfolio (and the stock market) hit their highest levels in early 2018.  This makes the recent period more extreme when viewed within the calendar year.   

The figure below shows a general simulation of Diamond since 1919, with the 1998 decline indicated by arrow “A.”  The green line is the model portfolio.  The brown line is the DJIA.  



Once can see several instances of large declines prior to 1998.  Even with these declines, the traded portfolio is superior to holding the DJIA over the long period of time.  Subscribers should understand that losses will occur. 

Performance of Diamond Compared to Alternatives

Diamond is the main model portfolio.  Up through early 2018, most subscribers were using it and this section focuses on it.  I also list other model portfolios for comparison.  All of the portfolios and performance figures are based in US dollars based on data supplied by Bloomberg.  CPM offers publications for investors in other countries.  As an aside, the best performing Focused 15 Investing model portfolios are those offered to Australian investors.  Please contact me for information about these model portfolios. 

Figure 1 below shows the performance for Diamond compared to alternatives that have been described in the past:
                The Dow Jones Industrial Average (DJIA)
                Vanguard’s Balanced Index Fund – VBINX
                The Russell LifePoints Growth Strategy - RALAX

Section A shows the performance of Diamond, along with two other model portfolios.  Mosaic (sg129) is included because it consists of all major sleeves used in other model portfolios and it was one of the original model portfolios in the publications.  It holds 8 ETFs.  I have also shown Zircon, which is just like Diamond but less aggressive.  Both Diamond and Zircon hold 3 ETFs.  

We will focus on the performance of Diamond.  In order to have a fairer comparison to other funds, we need to reduce the returns of the Focused 15 Investing model portfolios. This is because the performance for the other funds reflects actual costs associated with the management of those funds.  The average expense ratio for the alternative funds is 1.13%.  To be conservative, I subtracted 2.00% from the returns of Diamond for the comparison.  For example, the Diamond model portfolio returned 14.9%, annualized, over the 52-month period since the inception of the publication.  The return adjusted for costs is therefore 12.9%.  The variability of returns is not adjusted for costs. 

Figure 1


As one can see from the figure above, Diamond (adjusted for costs) returned -8.9% for 2018.  The DJIA returned -4.6%.  The worst performance among the alternatives is the Russell LifePoints fund (RALAX) at -8.1%. 

Over the roughly 52 months since the inception of the publication, Diamond (adjusted for costs) returned 12.9%, which is better than any of these alternatives.  The alternatives returned 1.9% to 9.6% over the 52-month period. 

The ratio of the return to variability is an important statistic.  The higher this number the better.  The ratio for Diamond (adjusted for costs) is 0.95.  The highest ratio of any of the alternatives is 0.71. 

Figure 2 below adds sections D and E.  Section D of the table shows the returns of alternative ETFs that do active asset allocation.  This information is from Bloomberg.  Diamond (adjusted for costs) is within the range of performance for 2018, but has higher returns and a higher ratio over the 52-month period than any members of this set of alternatives.    

Figure 2

Section E shows summary statistics for an additional set of alternatives.  These 114 funds that resulted from a search of the Bloomberg database.  The selection criteria targeted allocation funds (ETFs and mutual funds) and excluded private equity, money market, commodity and real estate funds.  The complete list of the 114 funds is shown in Appendix 1. 

The conclusions are similar across these different groups.  Diamond had a poor 2018, but its longer-term performance is still quite good compared to alternatives. 

Performance of the Range of Model Portfolios for 2018

This section reviews the performance of 14 model portfolios.  This section is detailed.  The bottom line is that most model portfolios performed better than their default mixes in 2018 and over the 52-month period. 

There are fourteen Focused 15 Investing model portfolios in two main publications.  Of the fourteen, only two had positive returns the year (12/29/2017 through 12/28/2018).  However, 11 of the 14 performed better than their default mixes.  The performance of the default mix is the return of ETFs in the model portfolio but held at constant weights over time.  All but one of the fourteen has less variability of returns than their defaults.  Less variability is a desirable feature.   

Figure 3 below shows the model portfolio returns for the fourteen main portfolios for 2018.  The publication for individuals has model portfolios with fewer ETFs for easier trading.  The publication for institutions holds portfolios that tend to hold more ETFs and fit commonly used asset classes.  Note that these figures are from the newsletter itself; an amount for fees has not been subtracted.

Column A shows the sleeve group number, shown as “sg’ on the weekly publication.  Column B shows the return of the model portfolio for the period.  Only two of the portfolios had positive returns (shown in green). Column C shows the variability of returns – the lower the number the better.  There are many numbers and I present a simpler graphical representation later. 

Figure 3 – 2018 Performance Information



Diamond returned -6.9% for the period.  Zircon returned -2.3%.  For comparison, the DJIA returned
-4.6%. 

A convenient way to understand performance is to review the return and variability of the model portfolio less the corresponding figures for the default mixes.  These figures are shown in columns F and G. 

In Figure 4 below, Columns D and E shows the return and variability of the default mixes for the different model portfolios for 2018. Our focus is on columns F and G, which show the difference between the traded and the default model portfolios.    

Figure 4 - 2018

For all but three of the fourteen model portfolios, returns were higher than their default mixes (column F).  For those underperforming their defaults (Diamond, Emerald, and Zircon) the underperformance with less than one-half of a percent for the period.  In all but one, the variability was lower than that of the default mix (column G).    

A few of the model portfolios had performance quite a bit better than their default mixes (Mosaic Plus sg213, Mosaic sg129, and sg 122).  All of these use what might be called a “slow-twitch” loss avoidance technique, which did not attempt to respond to the rapid return reversals that we experienced during late 2018.  I discuss this issue in greater detail later. 

The figures below show this information graphically.  The orange boxes indicate the return and variability of the default mix for the model portfolio.  The blue box indicates the return and variability for the traded portfolio.  The green arrows connect the model portfolio t its default.  The DJIA and Vanguard’s VBINX (60/40 Stock/Bond ETF) are shown for comparison; the names are listed in the table above.

Figure 5 





Ideally, we want the blue boxes to be very high and far to the left of the corresponding orange box.  That position would indicate that the Focused 15 Investing loss avoidance approach is working as intended.   For 2018, Diamond and Zircon had lower variability (a positive), but the returns were slightly less than the defaults (a negative). 

The figure below shows the broader list of model portfolios.  It shows only the sg numbers for clarity. 

Figure 6



This shows that for 2018, the traded portfolios generally had better performance characteristics than their default mixes.  Most of the blue boxes are to the left and higher than their default mixes (orange boxes) for the 2018 period. 


Model portfolio sg129 did very well in 2018 compared to its default.  The green line connecting the orange box representing the return and variability of the default mix and the model portfolio is the longest in the figure.  The model portfolio for sg129 is the second from the top. 

Figure 7 below is in the same format but shows performance since the inception of Focused 15 Investing on July 18, 2014.  


As indicated, these figures include the relatively poor performance of the 2018 period. I have indicated the position of Diamond and of sg129. While sg129 performed better than Diamond in 2018, its return for the 52-month period is not as high and it is more time consuming to trade (8 ETFs). For these reasons, many subscribers prefer Diamond.

The first page of the weekly publications shows the general performance expectations for the different model portfolios based on their returns since January of 2000. I consider these expectations to be attainable.

The only model portfolio that goes against the general trend on the figure above is sg200; its default mix is located at the lower left corner of the cluster. This is Onyx. Onyx is designed to be different from the others. It invests in four ETFs that have low variability, SHY (US 2y Bonds, 10y Bonds), XLU (Stocks of Utility companies), XLP (Stocks of Consumer Staples companies), and UST (US 10y Bonds x2). It shifts money into the ETFs expected to have higher returns. In contrast, all the other sleeves and model portfolios made from those sleeves invest in high return ETFs and then shifts away from them when they are expected to have low resilience. While Onyx’ performance has not been outstanding, I believe it may be useful in the future when interests rates are higher and move in a historically more normal manner. Thus, Onyx has a special role to play. I have two versions of Onyx, one with weekly trading and one with monthly trading. The monthly version is shown. I highlight Onyx because it might be useful for pairing with an aggressive version of Diamond. I will provide updates on this over the next few months.

Appendix 1 – Alternative Asset Allocation Funds (n=114)

RYDEX SERIES NOVA FUND-INV
KINETICS MARKET OPP-ADV A
WELLS FARGO DIV CPTL BLD-A
BOSTON TRUST ASSET MGMT FD
WALSER PORT AKTIEN USA
PLUMB BALANCED FUND
BRIDGES INVESTMENT FUND
STATE ST US EQTY VIS
VALUE LINE CPTL APPREC-INV
SIT BALANCED FUND
AMERICAN BALANCED FUND-A
JANUS HNDRSN BALANCED-S
BLCKRCK BALANCED CAPITAL-I
WF INDX ASST ALLOC-A
FIDELITY PURITAN FUND
VANGUARD WELLINGTON-INV
AMG CEP BALANCED-Z
VANGUARD BALANCED INDEX-INV
FIDELITY BALANCED FUND
USAA GRWTH & TAX STRAT
ABERDEEN INCOME BUILDER-INST
DODGE & COX BALANCED
LKCM BALANCED FUND
TOUCHSTONE BALANCED-A
GEORGE PUTNAM BALANCED-A
T ROWE PR PERS STRAT GRW
FIDELITY STR DVD & INC
PGIM BALANCED FUND-Z
EQUINOX AMPERSAND STRAT-A
TRANSAM M/M BALANCED-A
COLUMBIA BALANCED FUND-I
HARTFORD BALANCED HLS-IA
CALAMOS GROWTH & INCOME-C
T ROWE PRICE BALANCED FUND
MAIRS AND POWER BALANCED FD
T ROWE PR PERS STRAT BAL
TETON WESTWOOD BALANCED-AAA
DREYFUS BALANCED OPPORT-A
STATE FARM BALANCED FUND
EATON VANCE BALANCED FUND-A
GREEN CENTURY BALANCED-INV
TRIBUTARY BALANCED-INST
PIONEER CLASSIC BALANCED-A
JAN HND BAL-A US ACC
FIDELITY ASSET MANAGER 85%
FEDERATED MDT BALANCED-IS
HARTFORD BALANCED FUND-A
WF GROWTH BALANCED-ADM
ARCHER BALANCED FUND
ISHARES CORE AGGRESSIVE ALLO
HENNESSY BALANCED FUND
AMER CENT BAL-INV
AMER FNDS INC OF AMER-A
MFS TOTAL RETURN FUND-A
AMER CENT STR ALLOC AGG-INV
FDLTY ADV STR DVD & INC-C
CALVERT BALANCED FUND-A
FIDELITY ASSET MANAGER 70%
FIDELITY ADV ASST MGR 85%-A
JOHN HANCOCK BALANCED-A
CAVANAL HILL ACTIVE CORE-INV
FIDELITY ASSET MANAGER 60%
JPMORGAN DIVERSIFIED-L
FIDELITY ADV ASST MGR 70%-A
WESMARK BALANCED FUND
FRANKLIN BALANCED-A
FIDELITY ASSET MANAGER 50%
INVESCO EQUITY & INCOME-A
HOLLAND BALANCED FUND
SCHARF MULTI-ASSET OPP-INST
FIDELITY ADV ASST MGR 60%-A
PNC BALANCED ALLOCATION FD-I
PUTNAM DYN ASST ALL BAL-C
AMER CENT STR ALLOC MOD-INV
AMERICAN BEACON BALANCED-INS
FIDELITY ADV ASST MGR 50%-A
BRUCE FUND INC
BMO US DOLLAR MONTHLY IN-ANL
ARCHEA FD-NORDAM-B1
THRIVENT BALANCE INCOME PL-A
OAKMARK EQUITY & INCOME-INV
MAINSTAY BALANCED FUND-I
MANN & NAP PRO BLND MAX
FPA CRESCENT FUND
IVY BALANCED FUND-A
CORNERCAP BALANCED FUND
IRONCLAD MANAGED RISK FUND
BUFFALO FLEXIBLE INCOME FUND
CGM MUTUAL FUND
LK BALANCED-INS
DELAWARE FOUND MOD ALL-A
SATURNA SEXTANT CORE FUND
EATON VANCE HEDGED STOCK-A
GLDMN SCHS INC BLDR-A
USAA CORNERSTONE MODERATE
VIRTUS TACTICAL ALLOC-A
USAA CORNERSTONE MODER AGGR
MSIF GLOBAL STRATEGIST-A
GLDMN SCHS ALT PREMIA-A
MANNING & NAPIER INC-PB MO
1789 GROWTH & INCOME FUND-C
DAVIS APP & INCOME FUND-A
PALMER SQ STRAT CRDT-A
ICON RISK-MANAGED BALANCED-C
AMERICAFIRST TACT ALPH-A
DWS REAL ASSET-A
TOEWS TACTICAL MONUMENT
JAMES BAL GOLDN RAINBOW-RTL
EAS CROW POINT ALTERN-A
VILLERE BALANCED FUND
VIRTUS RAMPART M/A TRND-A
AMERICAFIRST QUANT STRAT-A
TOEWS TACTICAL OPPORTUNITY
CATALYST HEDGED FUT STRAT-A

1/09/2019

Is the Current Market Like 1987, 1998, or 2008? - Update #1

This post is the first update of my December 27th post, “Is the Current Market Like 1987, 1998, or 2008?”  While the key points of that post are described below, It can be found on my LinkedIn site or my blog

Large US stocks, as measured by the Dow Jones Industrial Average, became rated “1” as of December 28, 2018.  A rating of 1 means “slightly resilient” on a scale of 0 to 3, with 0 indicating “least resilient” and 3 indicating “most resilient.”  The shift to 1 from 0 had been expected. 

The DJIA had a rating of 0, “least resilient”, since November 30, 2018. This rating indicated that the index would likely decline dramatically with negative news and would be slow to recover.  Negative news of the day did indeed result in major price declines since the first of December.

This post discusses where US stock prices may go from here based on the status of CPM’s Market Resilience Indexes (MRI) as of January 4, 2019.  I discuss three scenarios mentioned in the original post that are relevant to current MRI conditions.  I present background information on the MRI framework at the end of this post.  Those unfamiliar with the MRI framework should review that information first. 

Current Status of DJIA – January 4, 2019

The current rating of 1 (slightly resilient) for the DJIA indicates that it has only a slight ability to shake off bad news. The Micro MRI (the shortest cycle) moved to the up-leg of its cycle and is at the 31st percentile of levels since 1918.  We can reasonably expect it to move through the 50th percentile and into the upper end of its normal range over the next several weeks.  Absent negative news, index prices can be expected to follow a similar path.

With only the Micro MRI providing resilience, the DJIA can be expected to move higher only briefly. When the Micro MRI comes to the end of its cycle in a few weeks, the DJIA index will again be rated 0 (least resilient).  Without any positive MRI, it will be most susceptible to negative news and price declines are likely. 

Thus, the most likely outcome right now is for the bottom of the US stock market to be in a month or so.  The bottom will be at a new low.  Basically, the current reading indicates that we are in a bear market rally.  This is described as Scenario #1, below.

Three scenarios are consistent with the current levels and directions of the MRI.  

Scenario #1 (reflects 2008) The bear market rally would move prices higher but not establish new highs. This would be followed by dramatically lower lows. The recent declines would be seen as an indicator of coming slower economic growth. The bear market rally would be like the one that began in February of 2008. During both February 2008 and now, the market had been sending signals of peak earnings (lower price-to-earnings but higher price-to-book ratios).  The market ultimately bottomed quite a bit lower in March of 2009. There may not be the excesses in the financial system that we had in the 2008-9 period, and we may not experience the same degree of loss that we experienced then.  Nonetheless, following the bear market rally, there could be another decline as the market adjusts to lower earnings growth rates.

Scenario #2 (reflects 1987) The current bear market rally is muted, but a longer-term positive price trend follows. The recent declines would ultimately be seen as a rapid, large-scale adjustment of valuations similar to the price declines of October 1987. There was not a strong rally (bear or otherwise) immediately after the October 1987 decline, but the decline did not foreshadow further deterioration in the real economy. Economic growth was strong pre-October 1987 just as it is now. In this scenario, the negative news catalysts for the price change (trade tensions, threats to Fed independence, and general Washington chaos) might have the effect of accelerating an adjustment for lower economic growth that would otherwise play out over a longer period.

Scenario #3 (reflects 1998) A fast recovery to price levels as high or higher than what has been recently experienced. The recent declines would be seen as being driven by negative news-of-the-day events occurring at a vulnerable time. This would be like the decline ending in August of 1998 related to the Long-Term Capital Management crisis, which was quickly resolved. Economic growth was strong in 1998 just as it is now. Then, as now, the Macro MRI indicated a negative trend in stock prices.  The LTCM crisis hit at a vulnerable time.  The 2015-2016 period was similar.  The Macro MRI turned negative in mid-2015, and the yuan devaluation scare produced price declines in August of 2015.  The US stock market remained vulnerable until mid-2016 and then became “most resilient” through the beginning of 2018.  I have called this a phantom bear market (link) because the period lacked resilience, but it did not have a significant negative event to catalyze a true bear market.  During this period, the Fed cited global weakness and held off raising rates, which seemed to accelerate the shift to a positive Macro MRI trend. 

Update as of January 4, 2019:
  • Price gains are, thus far, mild for a bear market rally. Stock prices are modestly off their recent low (evaluating weekly prices), yet the Micro MRI is already at the 31st percentile, indicating that the upside might be less than would be expected if the current level were lower, in the range of, say, the 5th to 10th percentile. This condition is consistent with scenarios #1 and #2. 
  • Powell, Yellen and Bernanke discussed consideration of global weakness in setting the path for rate hikes (American Economic Association, January 4, 2019). Trade tensions and a deterioration in confidence in Europe and Asia could accelerate a dovish shift in the Fed’s tone. While I do not see the Macro MRI shifting to its up-leg this week or next, this statement is like what was said and done in mid-2016. This would be consistent with scenario #3. 
  • World Stocks (MXWO) – Rated 1 (slightly resilient), with only the Micro MRI positive and moving higher. As of January 4, 2019, the Micro MRI is at the 64th percentile since 1975. Thus, it is roughly two-thirds of the way through a normal up-leg of its Micro MRI cycle. The peak may be just a few weeks away. While this index does not produce highly reliable signals like the DJIA, it does provide context. This is consistent with scenarios #1 and #2. 
  • Haven Assets (JPY, CHF, TY1) – This composite series is currently rated 3 (most resilient). That said, the Micro MRI is at a high level (the 86th percentile since 1987) and is likely to peak soon. When it does, there may be a short-term pause in the move higher of the haven assets, possibly during the bear market rally in the stock market. At the moment, it appears that the haven assets will move higher after that pause. This is consistent with scenarios #1 and #2.

Scenario #1 appears most consistent with the MRI conditions.  That said, some of the qualitative observations (which are not formally considered in the calculation of the MRI), such as weakening global economic growth leading to lower Fed rate increases, make scenario #3 more likely.  

Of course, the outlook can change over the next several weeks based on a change in the MRI conditions.  Based on market behavior over the last 100 years, for this brief move higher to be the beginning of a long-term trend higher, one or both of two indicators need to be present. 

First, the “Exceptional Macro,” the binary signal, would need to become positive.  Yet, it is not close to being present at this time.   If it does appear, it would be most consistent with Scenario #3.

Second, the Macro MRI itself would need to turn positive and do so without advanced warning from the Exceptional Macro.  This type of shift in the Macro MRI occurred in mid-2016 and a few times in the 1990s.  At the moment, it does not appear that this is likely to happen for the DJIA over the next few weeks.  If it does turn positive, it would be most consistent with Scenarios #2 and #3.

Please contact me with any questions or comments.

Jeffrey Hansen

Background on The Market Resilience Indexes (MRI)

The CPM Investing framework is organized around three main MRI that are generated each week for each index (e.g. DJIA, US 10y Treasury).
  1. The Macro MRI indicates the long-term trend of resilience and is indicative of the long-term trend in index price. Macro resilience increases and decreases with a cycle lasting several quarters or years. Each weekly reading indicates the level and direction of the long-term trend. The peak of the Macro MRI generally coincides with a peak in index prices. The cycles of the Macro MRI are only somewhat rhythmic; the cycles are often interrupted or truncated. 
  2. The “Exceptional Macro MRI” indicates when the Macro MRI is nearing the bottom of a cycle and is likely to turn positive. The onset of the Exceptional Macro serves as the most sensitive indicator of the bottom of a long-term market cycle.
  3. The Micro MRI indicates bursts of resilience typically lasting 6 to 24 weeks. The Micro MRI cycles are most apparent in the ups and downs of index prices throughout the year. They can be thought of as normalized price movements.

The MRI are additive. When they move together in one direction, they reinforce each other, and prices tend to follow that direction. When they move in opposition to each other, they tend to cancel each other out, and prices tend to be flat.

To describe the level of the Macro and Micro MRI, we indicate its percentile level within all its weekly levels since the inception of the index.  For example, the DJIA has over 5200 weeks of history since 1918. A level is described as, say, the 31st percentile within the historical range of levels.  Its direction is also important.  If it is moving higher (i.e., it is on its up-leg cycle), we say it is present and providing resilience.  If it is moving down (i.e., it is on its down-leg), we say it is absent and not providing resilience.  From readings of level and direction, we can estimate the MRI’s likely near-term course. 

For example, if a level of a Macro or a Micro MRI is currently low and moving down, we can guess that it will not move down much longer. As it nears the 1st percentile, the very lowest level of its historical range, we can expect the MRI to shift to the positive leg of its cycle and move higher. The Macro and Micro MRI are analog signals; levels move continuously and in a very general sine wave.   

The Exceptional Macro is a binary signal.  It is either present or not.  When it is present, it can overwhelm the analog Macro and Micro MRI to produce price gains even when the Macro and Micro MRI are absent. 

The ratings are defined by the number of positive MRI:
            3 = Most resilient
            2 = Moderately resilient
            1 = Slightly resilient
            0 = Least resilient

If all three MRI for an index are providing resilience, it is rated 3, “most resilient,” and prices may decline in response to bad news, but prices recover quicker and more completely. 

When none are providing resilience, it is rated “least resilient” and bad news produces bigger declines with slow and incomplete price recoveries.

The MRI levels and directions at a given time for different asset classes can help us position portfolios to favor the asset classes (e.g. stocks, bonds, and cash) rated most resilient and to avoid those rated least resilient. Trades based on MRI levels and directions are more reliable than trades based on actual price levels and directions. 

Note: For purposes of estimating stock market resilience, the Dow Jones Industrial Average is superior to other major indexes.  It has a 100-year history, has fewer constituents and indicates inflection points more distinctly, and has less interest rate sensitivity and commodity sensitivity than the Russell 1000 and the S&P 500.  Thus, while it sounds very old school to use the DJIA, it produces reliable signals and justifies the focus.  We do track and evaluate other indexes for context and confirmation.   

1/02/2019

Macro Trend Rotation

As of January 1, 2019, I am adding a new model portfolio to the publication. It is called Diamond-Zircon Rotation model portfolio. Over the last several months, many subscribers have shifted to Zircon from Diamond in response to my comments about the negative Macro (long-term) trend in stock prices. Both Diamond and Zircon are based on the same set of signals (which I call “D5”), but Diamond is more aggressive than Zircon; its gains and losses can be expected to be larger.

While the original intent of the Focused 15 Investing model portfolios was that each subscriber would select a model portfolio with a risk profile that is appropriate for their risk tolerance and time horizon, many subscribers have shifted model portfolios over time. This response from subscribers is perhaps inevitable. I have been open about the risks I see in the market and it is easy to rotate from one model portfolio to another. Given this usage of the publication, I sought to develop a more disciplined and robust way of rotating between Diamond and Zircon based on the larger market trends.

The Diamond-Zircon Rotation model portfolio uses a new signal, what I call the Macro Rotation Signal. To create the Macro Rotation Signal, I adapted a subset of the existing algorithms to focus on just the long-term price trend. By doing this, I was able to use algorithms that have been used real-time for about 10 years and have roughly 100 years of history.

The Macro Rotation Signal can be applied to various pairs of more aggressive and less aggressive model portfolios, like Diamond and Zircon. I am also reviewing the advantages of applying the Macro Rotation signal to Sapphire and Onyx (which consists of low volatility sector ETFs). 

The testing suggests that rotating between Diamond and Zircon can reduce weekly losses. Rotation provides the comfort factor of not using the aggressive ETFs (“DDM” in particular) in a negatively trending stock market.  It also enables a single model portfolio to have a wider range of risk exposures without having to trade a large number of ETFs. 

The result of this analysis is the Diamond-Zircon Rotation model portfolio (sg227). The individual existing Diamond and Zircon model portfolios have not changed and can still be used as before.

Now is a good time to introduce the Diamond-Zircon Rotation model portfolio because a key price trend measure (mentioned below) for the DJIA recently turned negative, which indicated that a shift to Zircon would be appropriate. I have spoken to subscribers to notify them about the availability of the Diamond-Zircon Rotation model portfolio.

For those following Diamond-Zircon Rotation, there may be some reduced return over the short-term should market prices begin a long-expected bear-market rally. Yet, the bottom of the current market cycle does not appear to be at hand and future market declines can be sufficiently large to justify the shift to the less-aggressive Zircon at this time.

The Diamond-Zircon Rotation model portfolio can be expected to produce slightly lower returns than Diamond over the long-term (by about 1.5 percentage points per year, annualized) but the variability of returns is also slightly lower. Again, a key advantage is reduced aggressiveness when the market is in a long-term vulnerable trend.

I tested the Macro Rotation Signal over three different time horizons. Because of data limitations, I could not do all tests over the longest time horizon. The first analysis covers 100 years of price history for the DJIA. This analysis focuses on the benefit of using the Macro Rotation Signals to move out of the market during times of vulnerability.

The Macro Rotation Signal is based primarily on two elements, a) the status on the Macro MRI, and b) the overall price trend of the DJIA using a technique similar to a widely used technical analysis tool called Moving Average Convergence Divergence (MACD). The MACD-related element is what turned negative over the last few weeks. The table below shows the performance statistics for the 100-year period for a simple model portfolio consisting of holding the DJIA in positively trending markets and holding cash in negatively trending markets.

These statistics are based on price change only; they do not include dividend return. Nor do they include any return for cash that might be held. The DJIA – Buy-and-Hold is the benchmark for this analysis; it is what would be achieved with no Macro Rotation.

(1918 – 2018)
Rate of Return (annualized, higher is better)
Variability (annualized standard deviation of weekly returns, lower is better)
Ratio of Rate of Return to Variability (higher is better)
DJIA – Buy-and-hold
5.81%
17.32%
0.34
DJIA using Macro Rotation Signal to get out of market and hold cash
8.49%
11.56%
0.73

Thus, the Macro Rotation Signal improves returns and reduces variability – both are positive changes. This result is seen in a higher Ratio of Rate of Return to Variability measure.

The average number of rotations per year is 1.8 over the 100 year period. Of course, these are not spread evenly over the period. Instead, they tend to be clustered. The figure below shows the DJIA price (log scale) in brown. The green line reflects the price change of our simplified rotation portfolio.


The better return for the simple model portfolio using the Macro Rotation Signal is based on identifying the negative trends in the market and avoiding losses. Subscribers may recognize this chart from the initial descriptions of the investment approach.

The following chart is of the same series but with a shorter time span, beginning November 1931, at the bottom of the stock market during the Great Depression. This time span highlights how Macro Rotation Signal works during different market environments.

The approach does best when there are market losses. It does not keep up with the buy-and-hold DJIA during strong trending markets, which are circled below.


Over this shortened time period, the following are the statistics for rate of Return / Return Variability (higher is better).


  • For the DJIA (buy and hold):       0.41
  • Using the Macro Rotation signal: 0.67


During the late stages of an ascending market the rotation signal produces a shallower slope than the buy and hold. This indicates that the Macro Rotation Signal reduces risk prematurely. Subscribers will recognize the tendency of the MRI-based approach to underperform in the late stages of an ascending market, which we see in the figure above. I adjust for this in the design of the model portfolios (e.g., Diamond, Zircon). The most challenging period marked “A” on the graph above.

The table below compares simulated returns for the Diamond-Zircon Rotation model portfolio to Diamond and Zircon over the period 1966 through the end of 2018, which includes period A indicated above. These simulations do not include dividend return or any return to the cash held in portfolios because of data limitations over this long time horizon.

1966 - 2018
Rate of Return (annualized, higher is better)
Variability (annualized standard deviation of weekly returns, lower is better)
Ratio of Rate of Return to Variability   (higher is better)
Simulated Diamond (D5 Signal Set using ETF “DDM”)
12.61%
14.88%
0.85
Simulated Diamond-Zircon Rotation
10.87%
12.71%
0.86
Simulated Zircon (D5 Signal Set using ETF “DIA”)
6.61%
7.44%
0.89
DJIA Buy-and-Hold
6.25%
15.89%
0.39

Zircon’s annualized rate of return is slightly higher than the DJIA Buy-and-hold, but the variability of returns is less than half. This boosts the ratio of return to variability to a high level.

One can see that the Diamond-Zircon Rotation model portfolio has much better performance than Zircon and slightly lower performance than Diamond. For many subscribers with longer time horizons, Diamond might be the right choice. Yet, for those with shorter time horizons or uncomfortable holding aggressive ETFs in negatively trending markets, the Diamond-Zircon Rotation portfolio might be the right choice.

On the three charts below, I show the Macro Rotation Signals for the period 1980 to the end of 2018. One can see the major tops and bottoms of the market, which are indicated. It is at these point that the Macro Rotation Signal adds value.



One can see from the figure some minor tops and bottoms. The Rotation signal is less likely to add value for these. This is particularly true during period A (from 1985 through 1994). Macro Rotation shows considerably greater effectiveness in the period of 2000 to 2018 than than during period of A. Going forward, we want to be prepared for many environments so it is important to consider the success of the approach during the difficult period A.

One can also see in the charts that some of the Macro Rotation shifts are clustered in time; a rotation in one direction is reversed just a few weeks later. To date, I have not identified how to reduce this clustering further without reducing returns.

Trading in the Diamond-Zircon Rotation portfolio is simple and, I believe, mitigates this issue. It holds only two ETFs, ETF “DIA” for the DJIA with no leverage. ETF “DDM” for the DJIA at two times leverage. It does not hold ETFs for long and short-term bonds. Instead, cash is simply help in one’s account. At this time, bond ETFs do not have high return so excluding them does not result in a big negative impact on returns.

As mentioned, I am reviewing applying the Macro Rotation Signal to other pairs of model portfolios. These might have superior performance characteristics while still being easy to implement.

Of course, subscribers are free to use any of the model portfolios. Please review the statistics and the various risks and select the model portfolio appropriate for you.

12/27/2018

Is the Current Market Like 1987, 1998, or 2008?


The DJIA continues to be rated “most vulnerable” with a rating of “0.” The scale is 0 to 3, with 3 indicating “most resilient.”  The DJIA achieved the most vulnerable rating on November 30, 2018.  This rating indicates that the index would likely decline with negative news and would recover slowly.  Negative news of the day resulted in major price declines since the first of December. This post discusses where prices may go from here based on the status of CPM’s Market Resilience Indexes (MRI).   

The CPM Investing framework consists of three main MRI for each index (e.g. DJIA, US 10y Treasury).  The Macro MRI indicates the long-term trend in index price.  The peak of the Macro MRI coincides (generally) with the peak in index prices.  The Macro MRI indicates the pricing cycles of resilience lasting several quarters or years. 

The Exceptional Macro MRI indicates when the Macro MRI is nearing a bottom of its cycle and is therefore likely to turn positive.  The onset of the Exceptional Macro serves as the most sensitive indicator of the bottom of a market. 

The Micro MRI indicates the bursts of resilience lasting typically 6 to 24 weeks, depending on environment and asset class.  The Micro MRI cycles are most easily seen in the ups and downs of index prices throughout the year.

The MRI are additive.  When they move together, all moving higher, for example, they reinforce each other and prices tend to follow that direction.  When they move in opposition to each other, some moving up and some moving down, they tend to cancel each other out, and prices tend to be flat. 

To describe the current level of an MRI, we indicate its percentile level within all its weekly levels since the inception of the index.  For example, the DJIA has over 5200 weeks of history since 1918.  The current level is described as the 10th percentile in its historical range.  From that reading, we can see that its level is toward the lowest level of its historical range.  If the level is currently moving down, we can guess that it will not move down much further.  When it is at or near the 1st percentile, it is at the very lowest level of its historical range.  Since the levels are mean reverting, we generally expect the MRI to shift to the positive leg of the cycle and move higher.  This cycle is relatively consistent over time. 

When the Macro and Micro MRI are moving higher (in the positive legs of their cycles), we say they are present and providing resilience.  If the Exceptional Macro is also present, the index has a rating of 3; all three MRI are providing resilience.  When bad news occurs, it may drive prices down.  But the strong resilience of the market will generally cause prices to recover quickly. 

When none are moving higher, the market has a rating of 0.  During these periods, bad news produces bigger declines and slow and incomplete recoveries. 

The MRI levels and directions at a given time for different asset classes can help us position portfolios to favor the asset classes (e.g. stock, bonds, and cash) most likely to be resilient and to avoid those most likely to be vulnerable. For professional investors, this information can add a timeliness element to their existing investment activities.

With its current rating of 0, the DJIA has little ability to shake off bad news.  That said, the Micro MRI (the shortest cycle) is at a low level (10th percentile since 1918) and is therefore poised to turn positive.  This implies the market is likely to soon display short-term resilience that could move the DJIA price higher from its current level.  An absence of bad news or a positive catalyst is often needed to initiate an up-leg of the Micro MRI cycle. 

Yet, when only the Micro is providing resilience, the index price will generally move higher only briefly.  Based on general market behavior over the last 100 years, for this brief move higher to be the beginning of a long-term trend higher, one or both of two conditions would need to be true.  First, the Exceptional Macro would need to be present.  It is not close to being present for the DJIA. Second, the Macro MRI itself would need to turn positive without the advanced warning made by the Exceptional Macro.  This is not likely to happen for the DJIA over the next several weeks.  Thus, I believe the bottom of the US Stock market is likely to be several months away. 

Other MRI-related observations about the current environment:

  • US 10y Treasury Future – It is currently rated 2 (somewhat resilient), with its Micro and Exceptional Macro being positive.  The Macro (long-term) MRI is at the 3rd percentile – a very low level – in the historical range established since 1983 and is close to moving into its up-leg. The important Exceptional Macro is positive and thus foreshadows a shift to a positive Macro MRI.  When the Macro MRI does turn positive, it would signal the bond prices are resilient and likely to move higher.
  • Dollar Index (DXY) – new rating of 1 (somewhat vulnerable to declines) last week, down from a rating of 3 the week before.  This is a rapid shift toward greater vulnerability and suggests a weaker dollar over the next several weeks. 
  • US 2y Yield – shifted to a rating of 0 (most vulnerable to declines) on 11/30/2018.  The Macro MRI abruptly turned negative on that date, having been at the 100th percentile (since 1976), obviously an extremely high level.   In contrast, the Micro MRI is at a very low level for this series – corresponding to the 2nd percentile over the same time period. While the Micro MRI is likely to become positive in the next few weeks and provide temporary support for the 2-year yield level, at the moment, the longer-term (Macro) trend for the 2-year is for lower yields. 
  • US 10y Yield – shifted to a rating of 0 (most vulnerable to declines) last Friday.  The Macro MRI abruptly turned negative last week (12/14/2018).  While the Micro is at a low level and will likely soon turn positive, it will then be rated only a 1, “somewhat vulnerable to declines.”
  • The following stock market indexes have ratings of 0 (most vulnerable to decline): UK Stocks, Europe stocks, Japan stocks, Emerging market stocks (MXEF), Shanghai Composite. Many of these have Micro MRI that are toward the lower ends of their historical ranges and could begin to experience a bear-market rally.  Yet, the Macro MRI is clearly moving in a negative direction from historically high levels, reinforcing the view that these rallies will be brief.  The Exceptional Macro is not close to being present for any of these indexes.  The exception to this generalization is that the Shanghai Composite seems a bit closer than the others to shifting to a more positive long-term trend.  Thus, the mid-term outlook for global stocks is quite negative.
  • Regarding inflation concerns, the relative leadership of global inflation-linked bonds vs. global nominal bonds had called for inflation-linked bonds to be favored (with a rating of 3, most resilient) as recently as November 23, 2018.  Now, inflation-linked bonds are less resilient than nominal bonds (the relative leadership series is rated 1, meaning avoid inflation-linked bonds and favor nominal bonds).  By this measure, a deterioration in inflation expectations has happened quickly. 

Many of these changes have occurred quickly compared to their historical norms.  This may be an additional negative sign for stock markets in general.  Recent negative news has come at a particularly vulnerable time and has sent prices lower. 

Thus, stock prices globally are likely to remain “most vulnerable to declines,” with a rating of 0, or “somewhat vulnerable to declines,” with a rating of 1, for several months.  If general historical precedents based on 100 years of history for the DJIA hold true for the current mix of MRI, any rally driven by a positive Micro MRI will be quickly followed by lower lows. 

While this is the most common scenario (identified as scenario #1 below), there are two scenarios in which the market has begun a long-term trend higher from a configuration of MRI levels and directions much like those we are now experiencing.   All three of these scenarios have levels and directions of MRI like the current state for the DJIA.

Scenario #1 – The upcoming bear-market rally would move prices higher but not establish new highs.  This would be followed by dramatically lower lows.  The recent declines would be seen as an indicator of coming slower economic growth (the market has been signaling peak earnings for the last eleven months). The bear-market rally would be like the one that began in February of 2008.  During February 2008 and now, the market had been sending signals of peak earnings (lower PER but higher PBR measures).  In the earlier period, the market ultimately bottomed quite a bit lower in March of 2009.  We may not experience the same degree of loss as that was experienced in the 2008-2009 period, but following the bear-market rally there could be another decline as the market adjusts gradually over several months to lower growth rates.  

Scenario #2 – The upcoming bear-market rally is absent or muted but a longer-term positive price trend follows.  The recent declines would ultimately be seen as a rapid, large-scale adjustment of valuations similar to the price declines of October 1987.  There was not a strong rally (bear or otherwise) after the October 1987 decline but the decline did not foreshadow further deterioration in the real economy.  Economic growth was strong pre-October-1987 just as it is now.  In this scenario, the negative news catalysts for the price change (trade tensions, threats to Fed Independence, and general Washington chaos) might have the effect of bringing forward in time an adjustment for lower economic growth that would otherwise play out over a longer period (as in scenario #1). 

Scenario #3 – A fast recovery to price levels higher than what has been recently experienced.  The recent declines would be seen as being driven by unfortunately-timed news-of-the-day events. This would be like the decline ending in August of 1998 related to the Long-Term Capital Management crisis, which was then quickly resolved.  But the DJIA price trend was much stronger in 1998 than it is today.  Economic growth was strong in 1998 just as it is now. 

Scenario #1 has the bleakest mid-term outlook. Scenarios #2 and #3 are more positive. For all three, an appearance of the Exceptional Macro, or a new positiver trend in the Macro MRI (within being foreshadowed by the appearance of the Exceptional Macro) will indicate subsequent market action. I will provide update on these scenarios over the next several weeks.

In the meantime, CPM research indicates that the prudent positioning of a multi-asset portfolio is to favor bonds and cash.  Avoid stocks.  The recent sharp declines may not have ended yet.  The algorithms for our more active portfolios will likely attempt to take advantage of any bear-market rally.  The algorithms for our less active portfolios are more likely to sit out the bear-market rally.  

Please contact CPM Investing with any questions or comments.