Please Note: Blog posts are not selected, edited or screened by Seeking Alpha editors.

Asset Allocation For All Markets

To establish an asset allocation approach that works in both bull and bear markets requires one to look beyond variability of returns. One first needs to accept and understand that capital market risk is endogenous   It requires an acute sense of secular bull and bear equity market history and the consequences of not being correctly allocated.   For as Peter Bernstein once said consequences are more important than probabilities. It necessitates an approach to primarily protect against portfolio losses in addition to safely growing assets when market trends turn positive. To begin one needs to look no further than the existing factors and approaches available to investors today.  
 
Unfortunately, many still relate to the 80’s and 90’s bull market in stocks when it comes to making asset allocation decisions. The most effective strategy is still believed by many to be a “buy and hold” strategy with 100% allocation in stocks if you have a time horizon greater than 10-15 years. This is what is considered to be both rationale and optimal for managing investment assets. The hard reality for any individual investing in stocks since the early 80’s, the first 18 years has been a fabulous time with substantially asset growth which was then unfortunately followed by decline to negative returns in equity assets over the last 9 years. These longer secular markets have been represented by periods in which the market moves in one overall direction while additionally being made up of shorter less pronounced market moves in an opposite direction. 
 
Here is the sequence of secular market cycles since 1950 represented by the S&P 500.
 
Bull Market: Jan ’50 – Dec ’72, 22 years 
Average Annual Return: 9.42%
 
Bear Market: Jan ’73 – July ’82, 9 years
Average Annual Return: 1.87%
 
Bull Market: Aug ’82 – Aug ‘00, 18 years 
Average Annual Return: 15.64%
 
Bear Market: Sept ’00 – July ’09, 9 years
Average Annual Return: -3.01%
 
Having a disciplined strategy of a 100% allocation to large company stocks such as the S&P 500 during a secular bull market has been the easiest approach to grow your assets. Unfortunately, this strategy would mean disaster for any retiree making conservative systematic withdrawals during retirement.   At a conservative 4% withdrawal rate (still considered to be the standard among certified financial planners) a retiree investing in the S&P 500 stock index via an open-end index mutual fund or an equivalent exchange traded fund (“ETF”) beginning in August of 2000, would have depleted 99%! of their total assets by the end of July ’09. 
 
Why are avoiding losses twice as important as capturing gains? The answer can be found in the arithmetic of losses and gains. Quite simply an investor living through a 20% loss in their portfolio can only make up that loss by achieving a subsequent gain of not 20% but 25%. As the loss grows beyond 40%, the need to achieve a subsequent gain grows even more as the table illustrates.   
 
 
Return
Portfolio Value
Return
Portfolio Value
Return
Portfolio Value
Return
Portfolio Value
Year 1
-20%
$80,000
-40%
$60,000
-60%
$40,000
-80%
$20,000
Year 2
25%
$100,000
67%
$100,000
150%
$100,000
400%
$100,000
 
Avoiding loss is commonly associated with conservative investments such as treasury bills and notes yet it is rarely considered when attempting to avoid inflation by making a long-term investment in stocks. Typically only the average historical return on stocks in considered when recommended to an investor as the best means of growing assets beyond the rate of inflation. The harsh reality for stock investors who experienced a 44% loss in stocks over the course of the 12 months ending Feb ’09, the pain caused by knowing one will need an 79% gain just to breakeven.   
 
Why have Wall Street firms and advisors not embraced a more dynamic approach to building investment products that work in both bull and bear markets? Undoubtedly because the majority of the investment community came to Wall Street primarily during the last bull market and still carry the buy and hold stock strategy dear to their hearts. Additionally, many still have yet to accept that secular bear markets exist and we have been in one since the stock market peak of 2000. With the realization that the buy and hold strategy only works in secular bull markets comes the necessity to develop a more innovative strategy to grow our investment assets in both secular bull and bear markets.
 
The obvious justification for asset allocation stems from the idea that a disciplined diversification strategy which harnesses the returns of various asset classes and balances risk measured through standard deviation is better than a strategy that jumps from asset class to another, according to whim, which may easily end up with worse results than a disciplined asset allocation strategy.
A fundamental justification for asset allocation is the notion that different asset classes offer returns that are not perfectly correlated, hence diversification reduces the overall risk in terms of the variability of returns for a given level of expected return. Therefore having a mixture of asset classes is more likely to meet the investor's goals.

Strategic asset allocation in practice originated from a 1986 academic paper by Brinson, Hood, and Beebower (NYSEMKT:BHB) on portfolio management and a diversification paper supporting Modern Portfolio Theory (MPT) from Harry Markowitz in 1952. Markowitz eventually won the Nobel prize for his paper which ultimately anchored MPT in the investment community as the most suitable and rationale approach to investing.

The real world asset allocation results from the 2008 economic meltdown paint a much different picture of asset allocation in which the only numbers that were truly moving higher were the correlations across asset classes. Between Oct and Nov of ’08 asset classes moved in sync and asset allocation portfolios dropped more than expected. The 12 month correlation for example between the Vanguard S&P 500 fund (ticker: VFINX) and the Vanguard Total Bond Market Index fund (ticker: VBMFX) jumped from negative 0.63 in Sept to positive 0.43 in October and 0.34 in November. Additionally, Gold, the asset class which in the past had provided a safe haven during stock market crashes had also become highly correlated with stocks and also fell in value. For example the 12 month correlation between the Vanguard S&P 500 fund (ticker: VFINX) and the Vanguard Precious Metals fund (ticker: VGPMX) whipsawed from negative 0.55 in June to positive 0.79 in October and 0.90 in November of ’08. Plus, the correlations with raw gold prices and the Vanguard S&P 500 Index fund went from negative 0.82 in August to positive 0.14 and 0.57 in Oct and Nov.

So the question for most asset allocation practitioners today is how do we change our approach to ensure that our asset allocation strategy protects us against the worst of market crashes where the traditional asset allocation approach was no protection from increased correlations. The answer and solution can be found in both old fashion charting and some new and old market sentiment factors sponsored by the Chicago Board Options Exchange.  

In Michael W. Covel’s best selling trading book “Trend Following,” the author provides in great detail identification and astounding performance histories of the most successful trend follower money managers whose basic strategies work in all markets. While the book may be light on behavioral finance theory the illustrations are like an anthropological study on the use of price direction as the single factor in making all investment decisions. This idea of charting price direction to identify trends also happens to sit at the heart of another investing approach called technical analysis detailed in a theory outlined in the Robert R. Prechter and Wayne D. Parker’s paper "The Financial/Economic Dichotomy in Social Behavioral Dynamics: The Socionomic Perspective." Socionomic theory is built on the idea that individuals easily make decisions through value propositions when it comes to buying a can of tuna but when it comes to buying stocks only one factor that matters most is price direction. This same notion of a herding mentality when it comes to investing in stocks was first introduced to the field of finance by Robert Shiller, and Nobel laureates Vernon Smith, Amos Tversky, and Daniel Kahneman.
So how does one avoid large market downturns? One simple trend following solution is detailed in a Journal of Wealth Management Spring ’07 paper “A Quantitative Approach to Tactical Asset Allocation” by Mebane T. Faber which uses the 10 month trailing return of equities to make tactical market timing decisions to get in and out of stocks. 
 
To support the use of the 10 month moving average, additional measures to identify market down trends include two indicators supported by the Chicago Board of Exchange (NASDAQ:CBOE): the S&P 500 VIX and the Put/Call Ratio.
 
The VIX was first introduced in a paper by Duke University professor Robert E. Whaley in 1993. From the CBOE website the VIX is defined as “a key measure of market expectations of near-term volatility conveyed by S&P 500 stock index option prices” whose goal is to measure implied volatility in stocks over the next 30 days. For stock market participants, the VIX has become popularly known as the “fear index” and is widely followed since it has been shown predominantly to have an inverse relationship to falling stock prices. To effectively test and use the VIX as a market timing tool it is necessary to smooth the VIX prices since they are as volatile as stock prices on any given day. Normalizing the price swings of the VIX is achieved by dividing the S&P 500 price buy the VIX price and calculating a 10 month moving average of the S&P 500/VIX factor to determine changes in the overall trend of the market. Below is an example of the calculation and the effective changes in allocation to a dynamic portfolio since the ’08 market peak of the S&P 500 in May. Keep in mind allocations reflect the calculation results of the S&P 500 and the VIX prices from the preceding month. 
 
You’ll notice that the change in the 10 month moving average of the S&P/VIX factor is driven by either a rise in the VIX and/or a decline in the S&P 500.
 
Date
S&P 500
CBOE VIX
S&P 500/VIX
10 Mo MA
Change
VFINX
VBMFX
4/1/2008
1385.59
20.79
66.65
64.44
-2.60
0%
100%
5/1/2008
1400.38
17.83
78.54
66.11
1.67
0%
100%
6/2/2008
1280
23.95
53.44
65.15
-0.96
50%
50%
7/1/2008
1267.38
22.94
55.25
62.19
-2.96
0%
100%
8/1/2008
1282.83
20.65
62.12
60.04
-2.15
0%
100%
9/2/2008
1166.36
39.39
29.61
56.53
-3.52
0%
100%
10/1/2008
968.75
59.89
16.18
51.62
-4.91
0%
100%
11/3/2008
896.24
55.28
16.21
47.98
-3.64
0%
100%
12/1/2008
903.25
40
22.58
45.22
-2.76
0%
100%
1/2/2009
825.88
44.84
18.42
41.90
-3.32
0%
100%
2/2/2009
735.09
46.35
15.86
36.82
-5.08
0%
100%
3/2/2009
797.87
44.14
18.08
30.77
-6.05
0%
100%
4/1/2009
872.81
36.5
23.91
27.82
-2.95
0%
100%
5/1/2009
919.14
28.92
31.78
25.48
-2.35
0%
100%
6/1/2009
919.32
26.35
34.89
22.75
-2.72
0%
100%
7/1/2009
987.48
25.92
38.10
23.60
0.85
0%
100%
 
The CBOE’s Put/Call Ratio (PCR) is the third factor which can be used to help protect the portfolio. This contrarian indicator captures the volume of the number of put buyers divided by the number of call buyers. It’s contrarian simply because it’s widely accepted that at extremes the number of buyers entering the market to purchase put and call options are usually wrong when it comes to timing the direction of the market. When the number of bearish put option buyers hits an extreme it has a majority of the time signaled a reverse in market direction to the upside while an extreme in call option buyers is a bearish signal. To test the use of the PCR using a 12 month moving average of the PCR and multiplying it by the S&P 500 will normalize the trend of the PCR. 
 
Below is an example of the calculation and the effective changes in allocation to a dynamic portfolio since the ’08 market peak of the S&P 500 in May. Keep in mind allocations reflect the calculation results of the S&P 500 and the PCR from the preceding month. 
 
You’ll notice that the change in the 12 month moving average of the S&P * PCR is driven by either a fall in the PCR and/or a decline in the S&P 500.
 
Date
S&P 500
PCR
S&P 500 * PCR
12 Mo MA
Change
VFINX
VBMFX
4/1/2008
1385.59
0.67
928.35
979.25
9.42
50%
50%
5/1/2008
1400.38
0.86
1204.33
1005.63
26.38
50%
50%
6/2/2008
1280
0.73
934.40
997.05
-8.58
50%
50%
7/1/2008
1267.38
0.68
861.82
982.77
-14.29
0%
100%
8/1/2008
1282.83
0.68
872.32
981.76
-1.01
0%
100%
9/2/2008
1166.36
1.00
1166.36
991.17
9.41
0%
100%
10/1/2008
968.75
0.73
707.19
961.01
-30.16
50%
50%
11/3/2008
896.24
0.62
555.67
933.26
-27.75
0%
100%
12/1/2008
903.25
0.6
541.95
908.67
-24.58
0%
100%
1/2/2009
825.88
0.64
528.56
863.12
-45.56
0%
100%
2/2/2009
735.09
0.73
536.62
820.23
-42.88
0%
100%
3/2/2009
797.87
0.73
582.45
785.00
-35.23
0%
100%
4/1/2009
872.81
0.77
672.06
763.64
-21.36
0%
100%
5/1/2009
919.14
0.7
643.40
716.90
-46.74
0%
100%
6/1/2009
919.32
0.75
689.49
696.49
-20.41
0%
100%
7/1/2009
987.48
0.88
868.98
697.09
0.60
0%
100%
 
Assuming a 3 factor asset protection strategy which moves the 50% target allocation from the Vanguard S&P 500 fund (ticker: VFINX) to the Vanguard Total Bond Market Index fund (ticker: VBMFX) when all three factors (10 month moving average of the S&P 500, 10 month moving average of the S&P/VIX, and 12 month moving average of the S&P 500 * CBOE Put/Call Ratio) are triggered yields above benchmark performance. For the period, September 1996 through July 2009, the benchmark portfolio annualized return was 5.41% (8.43 standard deviation) while the dynamic portfolio return was 9.34% (6.37 standard deviation).          
 
Our next opportunity provides an approach to increase portfolio returns in bull markets through an increased allocation to domestic small caps when the real Federal Funds Rate turns negative. The rationale is detailed and outlined in the working paper “Stock Prices, Firm Size, and Changes in the Federal Funds Rate Target,” by Hui Guo. The paper illustrates and explains how the Federal Funds Rate has a “greater impact on small firms than on big firms because small firms usually have less retained earnings and thus are more vulnerable to the adverse liquidity shocks.” The use of the Fed Funds Rate in allocating between large and small caps is fully detailed in Douglas Roberts book, Follow The Fed To Investment Success. When the Federal Funds Rate falls below the rate of inflation it is a clear signal that easy credit is available to small firms. Over the period February 1991 through July 2007, when the Real Fed Funds Rate was negative small outperformed large on average annualized basis of 5.18% with an 80% rate of frequency. Shifting the 50% target for the Vanguard S&P 500 fund to the Vanguard Small Cap Index fund (MUTF:NAESX) when the real Federal Funds Rate turns negative, the benchmark portfolio, 50% large and 50% small, returned 9.51% annualized while the dynamic portfolio return was 10.67%.
 
Analysts” ability to forecast growth rates on earnings is exceptionally bad. James Montier, in his book, Value Investing: Tools and Techniques for Intelligent Investment cites a study on US analysts’ forecasts which details an average error rate of 93% for 24 month forecasts, 47% average error rate for 12 month forecasts and makes one wonder what’s the point of using analyst earnings and equity price forecasts when it comes to making investment decisions. Montier explains that the dependence of both an expected equity risk premium and discount rate in developing a Dividend Discount Model (NYSEARCA:DDM) makes such a calculation wrought with problems. Additionally, Montier theorizes that analysts may be susceptible to confirmation bias by creating earnings and dividend estimates that are a reflection of their short-term views rather than the long-term experience of actual earnings and dividends. However he does detail his own positive experience with consensus analysts’ forecasts by stating that “I often find that the growth implied from, say, a reversed – engineered DDM model is close to the analysts’ long-term forecast.” Montier’s findings combined with the evidence that long-term earnings forecasts are reflected in stock prices detailed in Journal of Investing 2009 paper, "The Empirical Relationship Between Stock Prices and Long-Term Earnings," one can theorize that the use of consensus forecasts across sectors could be used to optimally capture earnings growth at a reasonable price (GARP). To construct a large cap domestic portfolio using a GARP approach and sector ETFs, I downloaded the historical Price to Earnings (P/E) data along with each sector ETFs historical forecasted earnings growth rates from Vanguard 10 sector ETFs. To design a portfolio allocation I first calculated the historical monthly PEG (P/E divided by Earnings Growth Rates) ratios across each sector, calculated an average PEG, determined each sectors PEG as a percent of the average, and weighted the sector allocation for the following month based on each sectors PEG in relation to the average. The rationale of the approach is to ensure that the portfolio is buying earnings growth at a reasonable price, diversification across all sectors and overweighting and underweighting based on fundamentals. Below is an example. As you can see, for Consumer Discretionary which has a PEG that is 8.39% lower than the average, results in an allocation that is 8.39% greater than the 10% average allocation per sector.  
 
PEG
Consumer Discret ETF
1.09
Consumer Stples ETF
1.97
Energy ETF
0.25
Financials ETF
0.73
Health Care ETF
2.10
Industrials ETF
0.89
Information Tech ETF
1.16
Materials ETF
0.52
Telecomm ETF
1.43
Utilities ETF
1.76
AVERAGE
1.19

 
AVG PEG - PEG/   AVG PEG
Consumer Discret ETF
8.39%
Consumer Stples ETF
-65.44%
Energy ETF
78.89%
Financials ETF
38.65%
Health Care ETF
-76.08%
Industrials ETF
24.90%
Information Tech ETF
2.74%
Materials ETF
56.04%
Telecomm ETF
-20.47%
Utilities ETF
-47.62%
Allocation %
Consumer Discret ETF
10.84
Consumer Stples ETF
3.46
Energy ETF
17.89
Financials ETF
13.87
Health Care ETF
2.39
Industrials ETF
12.49
Information Tech ETF
10.27
Materials ETF
15.60
Telecomm ETF
7.95
Utilities ETF
5.24
 
Testing the GARP portfolio against an equal sector weight benchmark portfolio, the annualized results over the period Sept ’04 through July ’09 are as follows.
 
Benchmark Portfolio Return: 0.15
Dynamic Portfolio Return: 1.70
 
Summary
 
The 1st step in designing a sound asset allocation strategy for all markets is to recognize that secular bull and bear markets exist and you must first and foremost have an approach to mitigate bear losses. Portfolio losses have the greatest impact on portfolio value and mitigating them should be your primary investment objective. Remember consequences are more important than probabilities. To grow your equity portfolio assets during bull market up trends, the real Federal Funds Rate and buying earnings growth at a reasonable price should assist you in managing and building your large and small cap assets.
 
References:
 
Akerlof, George A. and Robert J. Shiller.  Animal Spirits: How Human Psychology Drives the Economy, and Why It Matters for Global Capitalism (Princeton University Press, 2009).
 
Bernstein, Peter L. Against The Gods: The Remarkable Story of Risk (John Wiley and Sons, 1998).
 
Gary P. Brinson, L. Randolph Hood, and Gilbert L. Beebower (July/August 1986). “Determinants of Portfolio Performance”. The Financial Analysts Journal.

Joe Callaghan , Austin Murphy , Mohinder Parkash , Hong Qian (2009). “The Empirical Relationship Between Stock Prices and Long-Term Earnings”. The Journal of Investing.
Covel, Michael. Trend Following (Updated Edition): Learn to Make Millions in Up or Down Markets (FT Press; 1 Updated edition, February 2009).
 
Faber, Mebane T. (Spring 2007). “Quantitative Approach To Asset Allocation”. The Journal of Wealth Management.
 
Guo, Hui. (May 2005). “Stock Prices, Firm Size, and Changes in the Federal Funds Rate Target”. CFA Digest.
 
Kahneman, Daniel; Tversky, Amos (1979). "Prospect Theory: An Analysis of Decision under Risk". Econometrica 47.
 
Markowitz, Harry M. (March 1952). "Portfolio Selection". The Journal of Finance.
 
Montier, James. Value Investing: Tools and Techniques for Intelligent Investment (John Wiley and Sons, 2009).
 
Parker, Wayne D., and Robert R. Prechter Jr. (2006). "The Socionomic Theory of Finance and the Institution of Social Mood: Pareto and the Sociology of Instinct and Rationalization," presented at the meeting of the Association for Heterodox Economics, London, England, July 14-16, 2006.
 
Roberts, Douglas S. Follow The Fed To Investment Success (John Wiley and Sons, 2008).
 
Whaley, Richard E. (Spring 2009). "Understanding the VIX". The Journal of Portfolio Management .
 


Disclosure: 65% Short S&P 500, 35% cash