The Investment Philosophy
The philosophy is based on an observation that asset prices including stock markets move in cycles; they do not move in straight upward lines. The cycles are very irregular and can last from a few months to many years which makes them difficult to predict; nevertheless the cycles do exist (see research). The most lucrative "home run" trades are available around the major turning points in the cycle.
Second observation: there are periods of strong stock market performance (we call them "In Periods") that may be attributed to the macroeconomic environment favourable to corporate profits growth.
Third: the USA stock market is the largest and most correlated with its economy while smaller international markets tend to follow Wall Street disregarding their own economic fundamentals especially during the major corrections and US stock market crashes. So investors cannot hide from the US market falls by investing in the international equities and should focus their research effort on the most important USA market.
Fourth: stock picking strategies rarely outperform the market average over longer periods while they increase the costs and investment risk. Most individual stocks long term performance can be linked to the overall market or segment performance. There are of course many prominent examples of the outstanding stock market pickers who outperformed the market consistently over long periods of time.
Given the evidence, our preferred Investment Strategy is timing the most reputable USA stock market index Dow Jones Industrial Average supplemented by sector rotation and stock picking that is correlated with the cycle. In the cyclical environment investors should have long (buy) positions during the broad major cyclical market upswings and avoid or reduce equities exposure during the cyclical downturns. In some cases short positions during "Out Periods" may enhance the overall returns. Similar results can be achieved with timing S&P500 and any other broad US stock market index.
The strategy aims to improve the chance to outperform the market, while reducing transaction costs and risk by enhancing diversification and trying to avoid the weak periods and major market downturns.
The ideal investment vehicles for that are broad market Index Funds and Index Exchange Traded Funds ETFs and the key question is how to identify the major Buy/Sell turning points in the cycle.
This is where our proprietary methodology adds the key informational value: by aiming to be more accurate in identifying the major turning points of the cycle.
Market Index Timing (Market Picking) - investment strategy for the future
Written in March 2008:
The "market picking" methodology should be more valuable in the coming years because the mid-term US stock market performance is likely to be subdued - nothing like the boom years of the past 26 years or so that were largely driven by a falling interest rates environment.
The graph depicting Dow Jones Vs Interest Rates over the post war period tells the story: the first 28 years of the rising interest rates (red line) clearly depressed the Dow Jones performance (black line) up to early 1980s. The second half of falling interest rates resulted in a booming stock market from early 1980s to early 2000s. Other asset classes (property, bonds) experienced similar cyclical patterns although at different turning points.
The current (updated in March 2008) short overnight rates at 2%-3% are very low by historical standards and cannot fall much further (zero is the limit) - they can stay flat or rise which means a slow stock market into the mid-term future, similar to the first half on the graph.
The good news is that the stock market should be still cyclical and the only way to produce good returns in such flat and cyclical market is by timing it (or "picking the market"): buy and leverage on the upswing and switch to cash or other classes that make sense on the downturn. Traditional "Buy & Hold" approach that reproduces the market index performance should have lesser chance of producing good returns.
Click for large version of the chart
Stock market follows aggregate Corporate Profits
The proprietary Business Cycle Index developed by George Bijak is effectively a broad leading indicator of the major turning points (not all) in corporate profitability for the whole USA economy. The aggregate US Corporate Profits are calculated quarterly by the USA Government and published in the National Income and Product Accounts (NIPA) at www.bea.gov
The U.S. Corporate Profits have been historically highly (93%) correlated with the major diversified USA stock market indexes: Dow Jones Industrial Average (DJIA) and S&P 500. Hence, the Business Cycle Index may be useful leading indicator of the turning points of the broad USA stock market.
The indicator is not perfect and it does not always work but it seems to be a helpful tool when forming a view on a broad direction of the US market over longer periods measured in quarters and years.
The Corporate Profits for the whole US economy don't always move in the same direction or by the same magnitude as the profits reported by individual companies or even the DJIA or S&P 500 (refer to "Comparing NIPA Profits with S&P500 Profits" by Kenneth A. Petrick).
Validated Methodology
The methodology is not perfect. A macroeconomic model cannot give always right signals about a complex and unpredictable stock market that is also significantly influenced by random events, psychology and political decisions impacting on the macroeconomic fundamentals. The only certain thing about it is that one day it will produce wrong signal and professional investors using this general opinion should be always prepared for such event and manage their risk.
The proprietary methodology was validated over more than fifty years of historical data. The Business Cycle Index, first developed and published in 2004, proved to be a more accurate indicator of economic conditions that led to major turning points in the Corporate Profits and correlated with it broad stock market indexes.
The forecast aims to only specify an average broad direction of the US Profits and the stock market and does not even attempt to set target values by certain dates; this would be simply impossible. Macroeconomics cannot be as accurate science as physics or maths - the principle best captured by Isaac Newton after he lost a fortune in the stock market: " I can measure the motion of the heavenly bodies but not the madness of crowds".
The formula is useful for identifying the more lucrative periods of the broad stock market when most stocks perform well due to the favorable macroeconomic environment when stock picking is less important. Such periods called “IN the market Periods” represented around 30% of time, lasted historically on average 15 months and occurred on average every 4 years ranging from a few months to a few years and each cycle was different – see performance tables and graphs.
The "OUT of the market Periods" represented on average 70% of time, lasted around 3 years ranging from a few months to many long years. They were volatile times and 4 in 10 OUT Periods experienced a major stock market crash. Our quarterly forecast formula, while more useful for identifying the beginning and end of the OUT Period, is less accurate or almost has no information value during the OUT period from quarter to quarter. Patient investors will appreciate however, that at the end of the OUT Periods the overall returns were mostly in single digits around 2% on average and below cash rate - not worth the risk - even though the market may have had large unpredictable gains (and losses) on the way.
How is this possible?
We think it is not possible to forecast the economy and stock market in the short term on daily, weekly or monthly basis with consistent mathematical accuracy because economy and stock market are far too complex systems influenced by countless independent and unforeseeable forces (the famous saying by Mr Rumsfeld - "the known and unknown unknowns" - springs to mind here).
We also openly admit that our methodology is unable to forecast the direction of the market for 70% of time during what we call "Out Periods".
Statistical methods, Chaos, Random Walk, Efficient Market theories try to explain why it is impossible although some of the theories were eroded by succession of discoveries and anomalies (see for example Robert Shiller 2002 or Financial Times survey of professional financial analysts: two-thirds do not believe markets are efficient).
Any methodology and analysis based on statistical tools that tries to explain and forecast the stock market requires that the stock market performance should follow so called "normal" distribution or Gaussian bell-shape curve. In reality, the stock market performance does not follow the normal distribution and the fundamental condition of "normal distribution" is not met. That makes statistically based methodologies inaccurate or at best less reliable. Risk management based only on statistical tools is also an illusion for the same reasons.
Many professional investors supported by high caliber scientists lost fortunes trying to predict short term market using the most sophisticated statistical methods. This would not surprise Nassim Taleb - author of The Black Swan, who is used to 6 sigma events occurring much more frequently than it would be observed in "normal" distribution environment.
Fed's Chairman Ben Bernanke appreciated the inherent uncertainty of the economic forecast in his speech at Boston College in 2009: "...I'd like to offer a few thoughts today about the inherent unpredictability of our individual lives and how one might go about dealing with that reality. As an economist and policymaker, I have plenty of experience in trying to foretell the future, because policy decisions inevitably involve projections of how alternative policy choices will influence the future course of the economy. The Federal Reserve, therefore, devotes substantial resources to economic forecasting. Likewise, individual investors and businesses have strong financial incentives to try to anticipate how the economy will evolve. With so much at stake, you will not be surprised to know that, over the years, many very smart people have applied the most sophisticated statistical and modeling tools available to try to better divine the economic future. But the results, unfortunately, have more often than not been underwhelming. Like weather forecasters, economic forecasters must deal with a system that is extraordinarily complex, that is subject to random shocks, and about which our data and understanding will always be imperfect. In some ways, predicting the economy is even more difficult than forecasting the weather, because an economy is not made up of molecules whose behavior is subject to the laws of physics, but rather of human beings who are themselves thinking about the future and whose behavior may be influenced by the forecasts that they or others make. To be sure, historical relationships and regularities can help economists, as well as weather forecasters, gain some insight into the future, but these must be used with considerable caution and healthy skepticism."
The Business Cycle Investor model takes a different approach that is not based on statistical tools such as commonly used Capital Asset Pricing Models, measures of volatility, portfolio theories or finding a "curve fitting" formulas using artificial intelligence and mathematical methods.
Instead, the proprietary formula measures a state of the macroeconomic conditions that historically preceded the turning points in the average direction of the US Corporate Profits.
The methodology does not include current or future market valuation metrics or relative bond prices and yield curves which are effectively aggregated "votes" or "expectations" about the future. If it did, it would be trying to forecast the future by relying on another forecast. This is not the way to improve accuracy of predictions as the approach would increase discrepancy and risk by replicating and multiplying inevitable mistakes in the original forecast.
Our method uses only the current actual officially published values of a number of macroeconomic variables available at the time of making the forecast; it does not make any attempts to forecast these components or any other variables. The outcome does depend on the quality of the used numbers which are often retrospectively revised by the Government, hence are not quite reliable. The only conciliation is that the whole market is working off the same imperfect numbers.
The proprietary model does not attempt and cannot accurately predict the exact values of the Corporate Profits and Dow Jones index or any other economic indicator for any particular day during the forecast period and beyond. It only gives a guidance that the average direction of Corporate Profits and the stock market is more likely to change over the next few months.
There are unpredictable risks in any stock market forecast and there are no guarantees.
Market forecasting science is by definition unreliable; one can only try to improve the odds of being right. The market does not always behave in a rational way for starters and psychology and motivations of the individual decision makers and the investing masses plays a huge role at certain times.
Speculation and market sentiment often dominate over macroeconomic fundamentals although less so in the beginning of an uptrend. Warren Buffett captured the essence of the speculative behavior when commenting on bubbles: "..like most trends, a bubble is driven by fundamentals at the beginning and then speculation takes over. As the old saying goes, what the wise man does in the beginning, fools do in the end". Our "In Periods" are effectively the times when the bubble begins.
Unexpected spikes in oil prices is an example of a major unpredictable risk factor. The stock market reacts to it without delays, be it for a short time, even disregarding positive long term macroeconomic forces that normally stimulate the Corporate Profits.
And there are risks related to trying second guess what the decision makers will do with monetary and fiscal policy - the decisions fundamental to the macroeconomic environment even if they impact the market with a delay. The interest rate and fiscal decisions are not made by an "autopilot" and sometimes may go against the perceived macroeconomic logic and popular expectations. An example of such an unexpected devastating for the stock market decision, according to the Fed Chairman Ben Bernanke, was rising interest rates and tightening fiscal policy at the beginning of the 1930s Great Depression that led to a record unemployment and eventually to the Second World War.
Another uncontrollable events include natural disasters, wars, acts of terror on a large scale and other factors that can move psychology of the market and short term direction of the economy and the stock market with obvious negative consequences for investors.
In this game, one can only aim to improve the chances of being right and investors should be prepared to manage the inherent risk to survive through turbulent times and occasional financial tsunami.
Weather forecast illustration
A good illustration of the new methodology is a case of weather forecasting because climatic system is also very complex and influenced by many unpredictable forces.
Due to the complexity of the system, it is not possible to consistently forecast accurate temperature for the next day, week or month in a specific place. No wonder that weather forecast have rather miserable track record. Similarly, some stock traders and analysts are trying to guess short term individual stocks movements - needless to say with poor results given enough time; it is not much different to playing in a casino - the longer one plays the more chances one has to lose the lot; the house always wins.
However, a more reliable weather forecast is clearly possible if the analysts would only concentrate on the longer term average temperatures and turning points aligned with seasonal patterns. Knowing for example, that we are at the end of Spring we could be fairly sure that the average temperature over the next few Summer months should be higher. Sure, some days may be colder than others but the average temperature should go up in Summer.
So, if our weatherman was "less ambitious" and made in March a prediction that the average temperature in the USA in July should be higher than in March, he would have a high chance of being right; although still not 100%. The nature can sometimes produce abnormal weather patterns being such a complex system: summer months can be very cold or spring months can be very warm occasionally so a prudent investor should not bet her house on the prediction in any case.
The Business Cycle Investor approach is similar to the "less ambitious" weatherman, although the stock market is even more difficult to predict because business or economic cycles are less regular than the weather seasons; they are all over the place ranging from months to many long years before turning direction.
The proprietary model measures economic conditions every quarter to determine if this looks like the beginning of “Summer” or “Winter” environment for the aggregate US Corporate Profits and the US stock market. Sometimes the "Winter" "Out Period" can drag for years which can be frustrating for many investors who are wired to trade everyday in expectation to make money everyday in any kind of market because this is their job.
Historical data helped to establish a neutral level of the Business Cycle Index – an equivalent of Spring and Autumn seasons. When the index crosses above that level it is time to go "IN the market" (Buy) in anticipation of improved Corporate Profits growth, when it drops below the neutral level it is time to get "OUT of the market" (Sell) because Corporate Profits growth is expected to weaken. The exact value of the Business Cycle Index is therefore less relevant and only the times of crossing the empirical neutral line carry the key information value about the turning points - that is why this website presents only a "binary" version (IN or OUT of the market) of the Business Cycle Index on the performance charts.
The BUY/SELL signal will not necessarily be followed immediately by the Corporate Profits trend turn - it would be strange if it did. The signal is only our view there is a higher than random chance of the tide turning over the next few months.
The validated model is now being used to make quarterly outlooks for the Dow Jones Industrial Average index with reasonable results.
Applications
Institutional investors, Funds Managers and Investment Banks use the proprietary research for global portfolio asset allocation, proprietary trading, index timing and leveraged strategies decision support. It is one of the research tools only and should be supplemented with other qualitative, quantitative and risk management methodologies before making an investment decision - please read legal disclosure
