Use link to right to download - may take a few moments Download PDF Version - Entire Series

Multiple Asset Class Investing:
Using Investment Simulations To Help Manage Risk

By Chris Ciovacco
Ciovacco Capital Management
January 2008

The Limitations of Solely Using Historical Data

The historical analysis used in the development of the CCM Multiple Asset Class Approach to investing is very relevant since it draws on asset class correlations and investment performance over a 37 year period (1970-2006). However, it is also limited since we know the future will be different from the past. For example:

  • Unlike 2000-2002, what happens if the next bear market in stocks lasts five years instead of three years?
  • What happens if the next ten years have inflation rates well above the historical average?
  • What happens if we get abnormally low inflation adjusted returns during the first ten years of your retirement or investment program?
A money manager's greatest fear is to invest a client's hard earned money just prior to a prolonged period of poor performance for the general stock market.

The CCM Investment Simulator

The CCM Investment Simulator is a detailed investment allocation simulator which helps us better understand the realistic range of possible future outcomes, both favorable and unfavorable, for a particular asset allocation. The model simulates the future using over 75,000 historical records of returns, inflation, and taxes dating back 37 years (1970-2006).

Monte Carlo simulation is a method that estimates possible outcomes from a set of random variables by simulating a process a large number of times and observing the outcomes. In our case, the Monte Carlo simulator is a computerized technique, which is the basis for probabilistic risk analysis, which replicates real life occurrences by mathematically modeling projected events, such as investment cycles, annual returns, and annual inflation rates. Monte Carlo simulation uses pre-defined probability distributions of risk variables to perform random modeling over many simulations.

The CCM Simulator allows the investor to examine multiple future investment paths based on varied market conditions, annual rates of return, and inflation. A conventional financial projection, which might assume annual fixed inflation of 4% and fixed annual growth of 10%, cannot take into account the variability that occurs in the real world.

Simulator - Basic Input and Assumptions

Historical data was collected for our investments/asset classes (or meaningful proxies) from 1970 to 2006 (37 years). The period 1970-2006 was selected since it includes several different investment climates. Prior to running simulations, the investor's current portfolio size, net future deposits, and anticipated withdrawals are entered, along with an asset allocation. The basic logic of the CCM Investment Simulator is outlined below:

  • Based on a historical probability distribution, choose which period of time below will be most similar to the first year of investing:

    • 1970-1981: This period of higher inflation and slower economic growth helped coin the term "stagflation". It was a difficult time for both stock and bond investors. Stocks produced poor returns of -14.84% in 1973, -26.65% in 1974, -7.36% in 1977, and -5.09% in 1981. Commodity investments provided an alternative to help combat what was at sometimes double-digit inflation. Inflation ran at 12.20% in 1974, 13.31% in 1979, and 12.40% in 1980. If you think the returns above are painful, try adjusting them downward for inflation. The model refers to this period as Bear Market One.

    • 1982-1999: This period was market by low inflation, strong economic growth, and excellent stock market returns. It is often referred to as the greatest investment boom in history. Commodity investors, for the most part, did not fare well. The model refers to this period as Bull Market One.

    • 2000-2002: Technology bubble bursts, bear market in stocks, favorable period for commodities and bonds, low inflation. The World Trade Center attacks added to the bearish tone. Stocks returned -9.05% in 2000, -12.03% in 2001, and -22.17% in 2002. To give you an idea of how difficult it is to recover from serious investment losses, the NASDAQ is still roughly 50% below its March 2000 high of 5,048. Think about that - it is almost eight years later and a NASDAQ investor in March of 2000 would still have a 50% loss today. The model refers to this period as Bear Market Two.

    • 2003-2007: Bull markets in most asset classes, rapid expansion of credit and money supply contribute to a weakening U.S. Dollar. This is the current bull market, which officially started in October of 2002. It is highlighted by the Federal Reserve flooding the system with money and credit via interest rates being reduced to 40-year lows. Globalization has also played a major role during an unprecedented environment of synchronous global growth. The model refers to this period as Bull Market Two.

  • How long will the cycle chosen above last? The length of the next cycle is determined randomly using a probability distribution for each of the four cycles. The probability distributions are based on the actual length of the four cycles (1970-1981, 1982-1999, 2000-2002, and 2003-2006).

  • Simulate a future annual return based on the mean (MEAN) and standard deviation (STD DEV) of the selected allocation's actual historical return for each respective cycle. For example, the MEAN annual return from 1970 to 1981 for the S&P 500 was 8.36% and the STD DEV (a measure of the uncertainty of the MEAN return) was 18.51%. The simulator randomly assigns an annual return based on the actual historical profile for each of the four cycles.

  • Simulate the annual inflation rate based on the mean and standard deviation of actual inflation rates for each of the four cycles. When the model is in the BEAR MARKET ONE (1970-1981) cycle, the inflation profile (MEAN & STD DEV) for this specific period is used.

  • Continue to simulate annual returns and annual inflation rates until the length of the current cycle in years has been reached.

  • Begin the process over again by selecting which of the four cycles (1970-1981, 1982-1999, 2000-2002, and 2003-2006) will occur next.

The model simulates annual returns and inflation rates for up to 65 years. It can also take into account other important factors such as taxes, withdrawals, and the effect of inflation on your purchasing power.

In an effort to capture more realistic volatility in the simulator, 12-month rolling returns are used as input for both the S&P 500 and the CCM Base Allocation from 1995-2006. Annual returns are used in both cases from 1970-1994. This causes a slight variance in historical annual returns and standard deviations for both the S&P 500 and CCM Base Allocation when compared to calculations based solely on annual returns. Using 12-month rolling returns allows us to capture intra-year volatility that may not show up in an annual return.

Using the CCM Simulator to Assess Risk-Reward Profiles

Graphs 3 and 4 show the results of 100 simulations for both the CCM Base Allocation and the S&P 500 Index (a 100% stock portfolio). While the simulator runs for 65 years, only the first 10 years are shown to make it easier to discern the differences in the uncertainty between the outcomes. The top of page (Graph 3) shows the S&P 500 Index and bottom (Graph 4) shows the CCM Current Base Allocation. More conservative investors have a lower risk allocation and more aggressive investors have a more aggressive allocation than the allocation used in this example.

Graph 3

Monte Carlo Simulation For S&P 500 Index

Graph 4

Monte Carlo Simulation For Multiple Asset Class Investing

Some comments about Graphs 3 and 4 - Simulation Results:

  • The CCM Base Allocation, under simulated future conditions was able to post better average, median, and worst-case results than the S&P 500 index.

  • Showing there is no perfect way to invest, the investor in the S&P 500 may be compensated with a better best case outcome, which is in line with risk-reward investing. I think most will agree the overall CCM Base Allocation risk-reward profile is more desirable than the uncertainty and risk associated with the S&P 500 Index.

  • While the model can show purchasing power based on simulated future inflation, the results in Graphs 3 and 4 are not adjusted for inflation.

Simulating Real-World vs. Published Inflation

Since inflation is a significant long-term concern, the model can be set to simulate true inflation rates rather than published inflation rates. True inflation rates refer to what consumers actually experience in the checkout line. The model uses actual published inflation rates from 1970-1982 and adjusted historical inflation figures based on changes made to the CPI since 1982 (1983-2006). The adjusted figures are based on research by John Williams of Shadow Government Statistics. The U.S. government published an average inflation rate of 3.07% between 1983 and 2006. Over the same period, Williams' research concluded the real figures produce an average inflation rate of roughly 7.03%. Similarly, between 1970 and 2006, his research concluded the average inflation rate was roughly 7.24% vs. the published average of 4.68%. You can learn more using the links below:

Risks and Limitations to the CCM Approach

While the CCM approach to investing can reduce risk from a historical and simulated perspective, it can by no means eliminate risk. Examining historical returns and simulating future returns is beneficial, but both rely on historical correlations between asset classes which change over time. Several factors have contributed to an environment where stocks, bonds, and commodities have all performed well from 2003-2007. This is an unusual situation which points toward changing correlations between asset class price movements. As a result, the risks in the current market are most likely higher than the historical data suggests. As asset managers, we must be prepared to adjust to an ever changing investment landscape. Obviously, investing in the asset markets will be difficult going forward for all participants, including those who utilize the CCM Multiple Asset Class Approach and models. However, the concepts presented here should help investors improve their odds of protecting and growing their assets on an inflation-adjusted basis. All market based investment portfolios are subject to principal loss, including a multiple asset class portfolio.

The next segment will discuss how we rebalance asset allocations from time to time based on changing market conditions.

Chris Ciovacco
Ciovacco Capital Management

Atlanta Independent Money Management Atlanta

Chris Ciovacco is the Chief Investment Officer for Ciovacco Capital Management, LLC. More on the web at

All material presented herein is believed to be reliable but we cannot attest to its accuracy. The information contained herein (including historical prices or values) has been obtained from sources that Ciovacco Capital Management (CCM) considers to be reliable; however, CCM makes any representation as to, or accepts any responsibility or liability for, the accuracy or completeness of the information contained herein or any decision made or action taken by you or any third party in reliance upon the data. Some results are derived using historical estimations from available data. Investment recommendations may change and readers are urged to check with tax advisors before making any investment decisions. Opinions expressed in these reports may change without prior notice. This memorandum is based on information available to the public. No representation is made that it is accurate or complete. This memorandum is not an offer to buy or sell or a solicitation of an offer to buy or sell the securities mentioned. The investments discussed or recommended in this report may be unsuitable for investors depending on their specific investment objectives and financial position. Past performance is not necessarily a guide to future performance. The price or value of the investments to which this report relates, either directly or indirectly, may fall or rise against the interest of investors. All prices and yields contained in this report are subject to change without notice. This information is based on hypothetical assumptions and is intended for illustrative purposes only. PAST PERFORMANCE DOES NOT GUARANTEE FUTURE RESULTS.