Friday update

20 minutes to spare and time to update our blog.

On February 3rd when the Dow was down 300 points we sold our short positions in the Emerging Markets (EDZ) and Russell 2000 (TZA) as well as closing out our position in the 20+ leveraged Treasury Bond ETF (TMF) for some nice short term gains.   Those gains covered losses in what remaining equities we had and actually resulted in a net positive day for us.

And to defy a client who in jest wondered if I was practicing “voodoo” in their accounts by having their net worth rise on a big down day (Being Irish automatically excludes any Voodoo ability) I went back to the TZA hoping for continued downside in US Markets, it didn’t happen and we incurred a 7% loss, so there.

What have we done lately?   We continue to hold on to our municipal and taxable bond funds which account for approximately 40% of our assets.   I still believe, lurking out there in the future is a market sell off in the 10% to 20% range.   Eventually the S&P 500 has to touch the 200 day moving average which it hasn’t done since 2012.  2013 is the exception to the rule when it comes to revisiting the 200 day moving average.

However, new Fed chair Janet Yellen provided positive testimony to Congress about the state of the US economy and that stopped the selling last week.

We will be entering a short but strong seasonal period that should last into April.   Hence we took some cash off the sideline and added to our minimal stock positions.  But at most we’re only 50% invested in equities and 40% in bond funds, the remaining 10% is in cash.

Emerging markets have stabilized and that is providing strength to US large/multinational stocks which do quite a bit of business overseas.   This large stock strength is coming at some expense to the type of small stocks we prefer causing a bit of lag in our accounts versus the indices.   This is likely a short term phenomena.

While a market sell off can occur at most any time, seasonality still points to a peak in US stock prices in March/April 2014 where we can expect an approximate 6 months of net weakness.   So caution is still advised.

There will a very good fat pitch coming to us in the September/October time frame and that’s the pitch we want to be ready for.


No positions

Stocks for hell or high water? No way

Advisers who tell their clients to remain fully invested in stocks, hell or high water is offering systemically dangerous advice. @Jesse_Livermore

We couldn’t agree more!

But we live and work in Lyons Colorado, so “high water” is a poorly timed phrase.

Taking profits in SGOC

We are taking the profit in shares of SGOC after the stock erupted for another 35%+ gain this morning.   Stocks that go parabolic usually become very unstable when profit taking eventually takes over and we’d like to be out of the stock before that begins to happen.   As you can see by the chart it has made similar leaps before but it usually gives up about half the gain in short order.


No positions

Brad Pappas

The similarities: Moneyball and investing

It was a cold and snowy day yesterday which isn’t uncommon this time of year in the Rockies, actually many of the heaviest snowfalls of the year occur in April.  But last night I was able to see Moneyball which is movie starring Brad Pitt and based on the experiences of Billy Beane who’s the general manager of the Oakland A’s baseball team.

The movie captured the highs and lows of the A’s 2002 season in which they found themselves losing three key free agents to larger franchises and trying to find a way to cope with the realities of being a franchise with a $35 million dollar payroll competing with teams like the Yankees who at the time had a $120 million payroll.

Due primarily to  one of the strongest unions in the country: the MLB players union, a salary cap on team payrolls is nonexistent and has made for an unfair, unlevel playing field between large market teams (Yankees, Red Sox) and small market teams (Royals, A’s, Padres for example).

Billy had to find a way to replace his stars with players who could contribute enough to make up for the loss in production yet be cheap enough to fit within his payroll.  He simply could refit the roster with high priced free agents or trade for players with large contracts, there was no question he couldn’t expand his player payroll budget.

Baseball has long been a sport in which scouting has played a significant role.  Scouts were frequently older baseball men who used their experience to extrapolate what a young players potential could be.  The process is highly subjective and as in most professional sports, most of the prospects never make to the major leagues let alone be stars.  Billie was tired of hearing the old phrases “he’s built like a ballplayer, a five tool player, a smooth level swing, and he’ll improve with time”.  All Billy wanted to know was: “Can the kid get on base?”  If the kid can’t get on base, be it from a walk or a hit he’s of little use to the team and all the pat phrases from the scouts won’t mean a thing.

While visiting the offices of the Cleveland Indians in an attempt to make a trade his offer is shot down based on a mouth to ear relay of information to the Indians GM from an unlikely looking young man.   This young man was one of the first to use data analysis to formulate opinions on their young players.

This young man catches the eye of Billy Bean because is on the cutting edge of incorporating data or quantitative analysis (QA) to determine if the player had the potential to contribute.  The young man didn’t care if the player could even field his position he only cared about his on base percentage (OBP).

The use of quantitative analysis (QA) in baseball was the brainchild of Bill James.  Eventually the use of QA in baseball spread, even to the large market teams like the Red Sox who in fact are owned by John Henry who is a proponent of QA in managing his hedge fund.  However, to this day the issue of QA is still contested since baseball is an old school sport where change rarely occurs and jobs are entrenched.

But this isn’t really a post about baseball as much as it is a post on the advantages of quantitative analysis in many of life’s endeavors especially investing.

In place of the subjective “it has a great looking chart” / “the stock has a low PE and a 2% dividend and we expect it to move 20% higher this year!” is our proprietary RMHI model that allows us to backtest over a decade of data to determine which balance sheet profiles and stock selection formulas actually work in creating above average shareholder wealth.   Can this stock “get on base?”

In baseball the large market teams are captivated every year by the super stars that hit the free agent market just as investors are fascinated by the attention getting stocks such as Apple.  The past does not equal the future and while the Angels may feel adding Albert Pujols to their roster for an average of $25 million a year till 2021 is a good deal, you must consider he is a richly valued player who is peaking at age 32.  Will he be able to contribute at ages 38 to 41?   For every Ted Williams there are many more Manny Ramirez’s who were done at 38 leaving the team stuck with a dead money contract.

But investors miss the questions they should be asking themselves:  Is my chance of making an above average return on Apple (which sells at 17x 2012 eps, 4.4 times revenue and 6x book value) better going forward than shares of a stock selling 9x 2012 earnings, 0.46 times revenue and .75 times book value).  This is the essence of Moneyball or as a wise man once said to me “Price is what you pay and Value is what you get”.  FYI the unnamed stock is Voxx International, symbol VOXX.

And the Oakland A’s won their division in 2002 with a better record than the year before.



RMHI January 2011 client letter

After the successful investment returns in 2007 I began to spend considerable time developing a model that would assist is replicating the returns in future years.   This project turned into a three year immersion into Quantitative Analysis and investment strategy development which combines both stock selection and market timing in one package.    Anyone who knows me would probably agree I loathe hyperbole in the financial press but the back-testing and real time present day monitoring of the models results continue to be consistent.   The returns generated in individual stock portfolios since this past October are exciting and an example of its potential.  However, as in any case of investment modeling and strategy the standard warning of “Past performance is not a guarantee of future performance” is always true.

The model went into full time use for RMHI clients in October when the anticipated mid-term election rally (see 4th Quarter 2010 client letter) was emerging, the single largest rally in the four year Presidential term.  So far, the results speak for themselves.

The RMHI models give our accounts a distinct advantage over mutual funds or large pooled portfolios in several ways:

  • The model allows our portfolios to focus on the best 30-40 stocks our model identifies rather    than diluting accounts with 200 to 500 or more common in most mutual funds.   Included is, as always our screening for negative animal and environmental companies.
  • Market timing with Hedging is built into the model which is not generally present in SRI or mainstream mutual funds.
  • RMHI being a “small” investment management firm allows for a distinct advantage since not only does it allow for greater concentration of the best potential holdings but also for investments in smaller capitalized companies where larger funds must pass over due to size restrictions. As the studies below indicate “small” with a value bias tends to be relative outperformer.
  • By and large the best investment managers in the world are capable of generating returns in the 30% zone and they frequently use some form of investment modeling.   Our small size allows us to “be under the radar” which enables us to invest in small and micro cap stocks which provide higher rates of return* and which most mutual funds or hedge funds are unable to.

A portion of the initial foundation of research for the model can be attributed to James O’Shaughnessy exhaustive research in his book “What works on Wall Street” where his research suggested that value factors such as Price/Earnings, Price/Book, Price/Cash Flow and particularly Price/Sales have consistently exceeded their benchmark indices and offer better guidance in stock selection rather than many Growth oriented factors.  In other words, stocks that are valued cheaply tend to outperform their peers.

The model has two primary functions: Stock Selection and Market Timing.

Stock Selection:  Stock selection is based on the time tested method of identifying potential investment candidates based on three measures:

Low Price to Revenue with Low Price to Cash Flow:  In addition to O’Shaughnessy’s research there is a significant amount of research which has determined the effectiveness of value-based selection criteria:

Robert Shiller stated in 1984 that fundamental value was probably the most important determinant of future price expectations.

Labonishok, Shleifer and Vishny (1994) determined that Value consistently outperformed “glamour” aka Growth stocks regardless of size or business cycle.

Fama and French (2007) determined that Value stocks (with low ratios of price to book value) have higher average returns than growth stocks (high price-to-book ratios) see graph below.

The ability to invest in the “Smallest” “Value” stocks will provide our investors with a distinct advantage over large institutional investment managers and mutual funds.    As the chart from Eugene Fama reveals:  The ideal sweet spot for stock selection is the crossroads of “Smallest” company size and “Value” which outperforms “Biggest – Growth” by almost 300%.

Short Interest: Research shows that growth stocks are more heavily shorted than value stocks and that short sellers tend to be right. Asquith and Meulbroek (1995), Desai et al. (2001), and Dechow et al. (2001) provide evidence that more heavily shorted stocks tend to perform poorly. A reduction in short selling/interest over recent months indicates less bearishness and potential future price appreciation.

Price Momentum: Simply stated, stocks that are cheap based upon their balance sheet assets relative to the stock price tend to outperform over extended time periods.   But!  It’s not enough to have just a cheap stock….you need a cheap stock that is moving higher, otherwise the market could be rising but your cheap stocks are stuck in the mud and not making any progress.

Jagedeesh and Titman (1993) observed a pattern of price momentum whereby past winners tend to outperform past losers over the next three to twelve months.
Conclusion:  Value stocks and in particular Small Value stocks have provided a much better return historically than popular Growth Stocks.   Using the historical results and applied with the use of Quantitative Analysis a diversified portfolio built upon the RMHI model should improve the expected rate of return without a significant increase in volatility, resulting in a much better risk-reward than could be achieved without the model.

In the RMHI model approximately 5000 stocks are scanned daily and ranked based upon our formula.   Stocks selected for investment are generally ranked in the top 2% and are held until they fall out of the top 5% category.

Market Timing: The goal of the RMHI Market Timing module is not to attempt to identify average market downdrafts but to catch major market swings up or down which tend to be driven by earnings.   In addition, it’s impossible to anticipate news events that could drive the stock market lower such as 9/11 or the JFK assassination.  Such events are random and typically short-lived.  Eventually markets resume the path they were on before the random event.  In addition, I wanted to avoid unnecessary trading or frequent signals that could have us trading excessively.  Essentially we’re looking for the big moves and ignoring the minor moves that would create unnecessary trading and reduce returns.

There are many tools that can be utilized to aid in predicting future market direction.  The accuracy and predictive ability of an indicator can and do change.  One of the most accurate in the past decade has been the rate of increase or decrease in earnings expectations by analysts for the S&P 500 Index.  Markets have moved in the direction of earnings expectations quite closely for the past 20 years.

In the past 10 years the stock market has been especially vulnerable when earnings expectations for the S&P 500 begin to falter.

In the 1990’s Dr. Ed Yardeni, based on statements made by then Federal Reserve Chairman Alan Greenspan developed the “Fed Model” as a method of valuing stocks versus bonds.   Simply stated, the original Fed Model was a tool that assisted in determining whether stocks were a better value relative to bonds.

The financial theory was simple enough:  money would flow to the asset value which was relatively more attractive and away from the overvalued asset.  In reality, the Fed Model by itself was a relatively poor indicator by itself but the basic theory was a good foundation to start from.

From 2000 to mid 2002 the Fed model gave a good account for itself as the model determined that stocks were of poor relative value to bonds and stocks did endure a prolonged Bear market.

In an effort to bridge the gap from theory to actionable buy-sell signals I experimented with many alternative indicators (including Sentiment, Monetary Policy, simple moving averages for the stock market) but determined that the Consensus Earnings Estimate for the S&P 500 provided the best indicator when combined with the Fed model.

As the red line in both charts below show, risings estimates for the S&P 500 index tend to be associated with good returns for investors even when bonds are of better relative value to stocks.  But rising earnings estimates combined with an attractive stock to bond comparison as determined by the Fed Model foretold extremely strong returns.  And, declining earnings tend to be associated with declining returns as well, especially when equities were poor value relative to bonds.

Blending the Fed Model with current forward looking earnings estimates proved to be an accurate and reasonable combination of effective and actionable buy – sell signals for U.S. equities since 2001.

The chart below shows the timing mechanism of selling when earnings move below the 20 and 40 week moving average and buying when rising above the 20/40 week average.  In case you’re curious, the last sell date was June 2008.

Back Test Results of the RMHI Model:  The data and test results date back to March 2001.   The effects of compounded returns over the 10 year period are self evident which accounts for the accelerating appreciation in value (red line).   The shaded areas are periods when the Market Timing module indicated that risk was very high for stocks and Hedging of portfolios was in place.   Hedging consisted of selling 50% of the value of the portfolio of stocks and replacing them with the Proshares Ultra Inverse S&P 500 ETF (symbol SDS), creating a market neutral risk profile.

Annualized Rate of Return net of trading expenses 54.86%

Average Total number of positions

Total Return net of trading expenses 7139%
S&P 500 return 11.87%
Annual Turnover 241%
Maximum Drawdown -28.4%
Percentage Winners 52.77%
Sharpe Ratio 1.87
Standard Deviation Model 28.4% versus 26.19% S&P500

Model Returns by Year including trading expenses gross of management fees

2001* 2002 2003 2004 2005 2006 2007 2008 2009 2010
RMHI Model 106.96% 78.99% 158.43% 112.9% 21.46% 59.38% 2.02% 23.94% 72.51% 39.21%
S&P 500 -1.38% -23.37% 26.38% 8.99$ 3.00% 13.62% 3.53% -38.49% 23.45% 12.78%
Excess Return 108.34% 102.36% 132.05% 103.91% 18.45% 45.76% -1.51% 62.43% 49.06% 26.43%

Investors are typically loath to endure a cyclical Bear Market which can last for 6 to 9 months or more.  My hope is that the use of Hedging of portfolios during periods of predicted market weakness will incline antsy investors to stay put.

Frequently asked questions:

“If your model indicates risk is high and the chances of a big market selloff are large, why not sell off all your stocks and put 100% into the SDS?”     A very valid question, back testing this concept showed that volatility of the portfolio would be much larger without any stocks to offset the “SDS”.   Bear Markets tend to have some very strong rebound rallies or whipsaws which cut into the gains made on the SDS and make any investor nervous.  You could get lucky and sell at or near the bottom, that’s certainly possible since investor sentiment at bottoms is extreme.  My preference is for the less volatile strategy.

“Is there an aspect to the model that you’re not completely happy with?”  Yes, the market sell signal in 2008 was excellent but the buy signal, which required earnings to exceed the 20 week moving average was slow in my opinion.   The absolute bottom for most stocks was November 2008 but the model did not go into buy mode till May 2009.  Ideally the hedges should have been removed when sentiment was truly extreme in November and slowly adding stocks afterwards.

“Is this the only model you have developed?  Are there others in case this one loses effectiveness?  Yes, in addition to the present model there are at least three others that I continue to monitor closely.  However, models can run hot or cold from one year to another.  I gave special preference for long term consistency which is why I’m using the present RMHI model.

“No model is perfect, what do you consider your models biggest weakness?”   There remains the risk of annual short term draw-downs or pullbacks in the portfolio.  I wish those could be smoothed but it’s not realistic at present and trying to do so can severely impair returns.  In an effort to temper this risk I’m considering employing the Ned Davis annual cycle chart as a roadmap.   It is predicting the start of a Bear Market in equities beginning in August 2011.

“Have there been any extended time periods where you believe the model would not have been effective?”  Yes, but this based upon experience rather than data.   The late 1990’s when the mania for Growth stocks, particularly Technology stocks was a rough time for Value stocks in general.  However the pendulum swung back sharply in the early 2000’s and the normal outperformance of Value reinstated itself.

“Why not simply sell all the stocks and just hold cash instead?’   This is another “all or nothing” approach which has significant risk in terms of “Opportunity Cost” or what you could have made had you held on to the portfolio with hedging.   The chart below shows this option:

Using cash in lieu of the SDS Hedge drops the annualized rate of return to 48.5% but the real cost is the impact on the compounding when compared to the chart using the hedge.

To sum it up, I’ve attempted to be as comprehensive as possible with this presentation by detailing the academic and data research that is the foundation for the model.   Some aspects such as Momentum and the weightings of the model elements must remain proprietary.  No model is perfect but based on everything I’ve monitored to date, the model does work effectively.  There are also the normal risks associated with any equity investments in particular surprise events.  However, one very important lesson that must be acknowledged is that subjective opinions will, in general hurts returns.  Maintaining discipline is essential to the performance of the model and I will maintain the effort to do so.

Disclosure regarding the SDS: Each Short or Ultra ProShares ETF seeks a return that is either 300%, 200%, -100%, -200% or -300% of the return of an index or other benchmark (target) for a single day. Due to the compounding of daily returns, ProShares’ returns over periods other than one day will likely differ in amount and possibly direction from the target return for the same period. Investors should monitor their ProShares holdings consistent with their strategies, as frequently as daily. For more on correlation, leverage and other risks, please read the prospectus.

Investing involves risk, including the possible loss of principal. ProShares are non-diversified and entail certain risks, including risk associated with the use of derivatives (futures contracts, options, forward contracts, swap agreements and similar instruments), imperfect benchmark correlation, leverage and market price variance, all of which can increase volatility and decrease performance. There is no guarantee that any ProShares ETF will achieve its investment objective. Please see the prospectus available at for a more complete description of these risks.