Despite Bad Mistakes, 75/50 Strategies Bouncing Back |
Date: Tuesday, February 3, 2009
Author: John Serrapere, Index Universe.com
In this column, we'll review the performance of a portfolio model designed to use both short- and long-allocation strategies for managing private wealth.
We'll call this our AI 75-50 Portfolio. (Excuse me, this isn't overt self-advertising but simply an easy way to reference the model from all of our records at my research firm, Arrow Insights). The portfolio attempts to capture 75% of the S&P 500's upside and 50% of its negative returns over 12-month periods. This performance profile drives the model's strategic allocation and tactical trades.
The 75-50 model may or may not experience lower risk than the market. It's definitely not appropriate for investors with time horizons that are less than three years.
The 75-50 Portfolio satisfies a need to employ capital originally allocated to hedge funds into a proxy. This need resulted from my displeasure with hedge fund managers, which lead to the pursuit of an alternative with clear performance objectives but without 2% management plus 20% incentive fees (2 & 20). It also leads to holding investments with limited transparency, low liquidity and business structure risks.
Future Active Indexer columns will report monthly performance updates and trade ledgers. A monthly format will be shorter in narrative with more attention paid to allocations and trades. This installment will be longer than forthcoming ones because we are bringing readers up to date on current views and portfolio moves made since March 2008. At this time, we will review the origin of our defensive posture, the universe from which we trade, and how we maintain beta - non-beta diversification in this age of heightened default risk.
Portfolio Exposures & Convictions
Figure 1 displays examples of how exposures are expressed when we have low, medium, and high conviction in asset price direction and valuations. Maximum conviction results in no more than a 170% gross position. In the example below, we are 110% long, 60% short, resulting in 50% net long. Once volatility reverts to normal ranges near 15% to 20% in annual standard deviations, I could see being 70% long with no shorts while holding 30% cash. This would be a time of low conviction on price direction while most assets are cheap. Most of the time, a medium conviction is expressed near 130% gross and 70% net long exposures. Currently, we hold moderately high convictions to trends and valuations resulting in a 144% gross exposure.
Our Bear Market Offers Us Abundant Opportunities & Extreme Risk
My December 30, 2008 interview by Murray Coleman, managing editor of Index Universe, reported that since the October 10 (S&P 839.80) and November 21 (741) market lows, AI 75, 50 was accumulating long positions for The Long Haul (10-20 years) because equity and bond were both pricing in a high probability of 7% to 9% annualized returns respectively. Here is commentary from the interview:
Active Indexer often interjects real-time views that support allocations and trades. Below are some excerpts from a message dated December 3, 2008 and titled: A Brief Outlook. It shows corporate bonds holding both high-reward and extreme risk. We are not naive.
Employment is a leading indicator, its decline there will be exasperated by the pulling of consumer credit, which might contract by -2bb in 2009 via FAS 140 and FIN 46R. Employment and consumer credit are contracting faster than at any other time since their data lives.
These (Economic) precedents imply via their historical behavior that we will have a depression, which has been defined as an economic contraction greater than -10% but less than -20% and that we are only about half way to one third of the way through our contraction.[1]
Credit spreads also imply that default rates will exceed the peaks seen in 1991 and the 1930s, which were near 13% and 16%. The current default rate is less than 4% so if it climbs to about 15% then losses from CDS and credit extensions will cause $100s of billions in more losses at financial firms. States & local govts and pension plans will also most likely experience more than $150 bb in additional losses.
The above supports Goldman's Sach's revision of S&P 2009 earnings to $65, which is close my $67 estimate made in Panic P/Es. I also found that during the above histories that the median 12-month trailing equity market P/Es stabilized near 11 but that it often fell to about 8 briefly. The above evidence supports potential market bottoms in the 520 - 737 range. That said the above Bears, experienced + 30% (S&P 998) to +40% (S&P 1075) counter trend rallies. Such a rally would most likely occur near year-end through the first couple of trading days in 2009.
You see the very short-term trend is down through Dec 8-15 and the primary trend is bearish. We need to be nimble because as we saw last week the Bear's counter trend rallies are strong so, we cannot rule out a counter-trend rally to near an S&P price near 1000 near year-end.
The fiscal stimulus and other government actions will affect the above outcomes. Their interventions should prevent an economic contraction greater than -20% (Great Depression II). We have to weigh the evidence as it unfolds. Most of all, we need to follow the market's price trends, which were pricing in recession early in 2008."
From this message, it is clear that during the December 30 interview, I was not clear about future corporate bond defaults. The message that I wanted to convey is that Default rates will soar in 2009 but it is unlikely that investors will lose money if they bought at the deep discounts available in November and December 2008. It is also more probable than not that real gross domestic product (RGDP) will fall more than -10% in 2009. So, hold your nose when buying low quality assets, but have comfort if price-yield are in balance with risk-reward.[2]
My December 2008 outlook called for tactically hedging long positions seen as profiting from a re-inflation of goods, services, and asset prices. The Fed's recipe includes the devaluation of the US Dollar to reduce our debt load. Further nationalization of our financial sector is also likely.
We are accumulating low quality assets priced for high current yield while also buying those that will appreciate more than more traditional allocations during times when our identified themes dominate. We do not expect normal real GDP growth, near 3% yearly for at least a couple of years, which may not materialize until sometime near the next Presidential Election in 2012.
In spite of the risks, we are buying assets that are compensating us for risk, as I indicted in the December 30 interview:
Cardinal Sins
Before reviewing AI 75-50 performance and asset positions (Figures 8-11), let's be frank. We made many bad trades in 2008. The best course would have been to have stuck to our February 2008 allocations and while timely covering our shorts during the fourth quarter of 2008. We did cover all shorts by November 21, but we were too smart for our own good. Our worst moves were in the spring of 2008. We sold all of SHY (1-3 Year Treasury Notes), consumer stables (FDFAX), Vanguard Health Care (VHT), and Vanguard Utilities (VHU). We then employed the proceeds to buy Energy Income & Growth Fund (FEN) and Macquarie Infrastructure (MIC). This was a Cardinal Sin. You try never to sell defense and then buy offense at prices that are too high. Our sins feed our worst month-end drawdown of -17.5% from February 2008 through October 2008.
AI 75-50 Portfolio Performance
The last two months of 2008 and January 2009 were the best three consecutive monthly returns since portfolio inception on March 19, 2004. AI 75-50 is up 14.5% on a net basis since October 31, 2008. 2009 has been great so far especially when compared to the S&P 500 and a traditional portfolio that typically allocates 60% to domestic stocks and 40% to domestic bonds. They were down -8.6% and -5.9% respectively while we were up 4%. The 2 & 20 crowd (HFRX) finally shed some Beta and posted a 1.3% positive return in January 2009.
After reviewing Hedge Fund Research's (HFR) Investable (open for new investors, X) Indices and The Barclays CTAs Investable BTOP 50 Index (top 50 commodity trading advisors ranked by assets under management), a peer group benchmark was designed. Our peer group is represented by 20%, 30% and 50% allocations to the BTOP 50 Index, the HFRX Absolute Return Index and the HFRX Equity Hedge Index. AI 75 -50 attempts to maintain a balance of source returns that expressed by our peer group composite.
For the full year in 2008, a -2.2% return bests all benchmarks and peers. Since the month-end of October 2007, AI 75-50 is up 2.8% through January 2009, which scores an opportunity gain of 46.7% and 28.8% relative to the S&P and a traditional portfolio. Hedge fund returns through January 29 show opportunity gains of 26.2% and 18.6% compared to HFRX and Peers (Figure 2).
Figure 3 summarizes results relative to the S&P 500 and The HFRX Global Hedge Fund Index. Cumulatively, AI 75-50 gained 48.9%, while the former benchmarks lost -20.7% and -7.9% since inception on March 19, 2004 through January 2009. Although, monthly standard deviations were the same at 2.5% for the portfolio and HFRX, volatility has been about 35% less then realized by the S&P 500 on a monthly and annual basis (ASD) while ASDs were about 20% greater than for HFRX.
Effective risk management requires that we fix a diligent eye on portfolio positions and all benchmarks. Figure 4breaks out our Peer Group into its parts so we can evaluate AI 75-50 relative them along with other benchmarks. Here performance is reviewed since inception, for the past 3 years and for the latest 12-month period through January 2009.
In Figure 5, we determine if the sum is greater than its parts, which is the basis of MPT analysis with Minsky, Chris Whalen (http://www.institutionalriskanalytics.com/), Eric Janszen (http://www.itulip.com/), Ben (the Pittsburgh broker from my early training) and other credit conscious teachers thrown in for real world optimization.
In the next column, we will review why it would have been best to do less in 2009, and we will cover our five best and five worst trades. We also will revisit Panic P/Es, technical charts, and our expected returns for our positions. We also will introduce my Price Index Score (PIS) and other proprietary asset pricing and economic indicators employed with MPT but Minsky in mind.
Endnotes
1. John L. Williams, Shadow Government Statistics, SGS Newsletter No. 47 of November 14, 2008.
19th Century style depressions were real economic contractions greater than 10%. Only the 1930s contraction is known as a Great Depression because at 26% it has been the only contraction greater than 20%.
2. SGS Newsletter No. 48 of January 3, 2009.
By April 2009, the recession will be timed as the longest since the Great Depression (its current 13 months ties the length of the second-dip of the Great Depression in that late 1930s), where, at present, its length was exceeded by the 1973 to 1975 recession and by the second leg of the double-dip recession of the early 1980s
Reproduction in whole or in part without permission is prohibited.
