Skip to content

A Quantitative Approach to Tactical Asset Allocation (Faber): summary and key takeaways

Faber tests a monthly rule that holds an asset only when its price is above its 10-month simple moving average, otherwise cash. Applied to the S&P 500 since 1901 and to a five-asset portfolio since 1973, it lowers volatility and drawdown. The 2013 update adds the 2006-2012 out-of-sample period.

Faber, M.T. · 2007 · The Journal of Wealth ManagementSummary updated 20 November 2026
Published2007Journal article
Asset focusMulti-assetTAA
TypeBothTactical Allocation
Reading time4 minthis summary

A Quantitative Approach to Tactical Asset Allocation (Faber) at a glance

The question
Can a simple price-only trend rule, applied the same way to every asset class, reduce risk in a global multi-asset portfolio without hurting returns?
Dataset & period
S&P 500 total return, 1901-2012, monthly (pre-1971 series constructed by Global Financial Data). Five-asset portfolio (S&P 500, MSCI EAFE, US 10-year government bonds, GSCI commodities, NAREIT REITs), 1973-2012, monthly. Out-of-sample window: 2006-2012. Extension tests use thirteen asset classes.
Method
At each month-end, the model buys an asset if its price is above the 10-month simple moving average. If the price is below, the asset's slice moves to Treasury bills. The same rule and parameter apply to every asset. The GTAA portfolio splits capital equally across five asset classes, and each slice is either invested or in cash. Taxes, commissions and slippage are excluded from the backtests.

Key findings of A Quantitative Approach to Tactical Asset Allocation (Faber)

  • On the S&P 500 since 1901, timing earned a compound return of 10.18% versus 9.32% for buy and hold, with the system invested about 70% of the time.
  • In the late 1920s and early 1930s bear market, timing cut the drawdown from 83.66% to 42.24%.
  • In the five-asset GTAA portfolio (1973-2012), drawdown fell from 46% to less than 10%, and volatility fell to single digits.
  • In the 2006-2012 out-of-sample period, GTAA beat buy and hold in only three out of seven years but ahead by over two percentage points per year, with lower volatility and drawdown.
  • Across the five markets, returns were on average 60% lower and volatility 30% higher when price was below the 10-month moving average. Commodities were the exception on volatility.
  • Holding 10-year government bonds instead of T-bills as the cash leg added 1.37% per year for marginally more volatility and drawdown.

Robustness, as reported in the paper

  • Moving average lengths from 3 to 12 months were tested on 1973-2012 and all behaved similarly (the chart itself is not in the extracted text).
  • Results after 2005 are treated as out of sample, since the original paper was published in 2006 with data through 2005.
  • The bond-cash variant was also checked over 1973-1981, a period of sharply rising interest rates, and the extra duration still helped.
  • By decade, the S&P 500 timing model improved drawdown in all but two decades and improved risk-adjusted returns in about two-thirds of decades.

Caveats and limits

  • Taxes, commissions and slippage are excluded from the backtest. The paper says taxes are a real issue because turnover is almost 70%, versus about 20% for a typical buy and hold portfolio.
  • The paper warns that a portfolio's largest drawdown is always in the future and says a 20% drawdown is a plausible scenario, despite the historical figure of about 10%.
  • Trend following can lag buy and hold in strong bull markets such as the 1990s and may suffer whipsaws in oscillating markets, per the paper.
  • Most headline statistics (CAGR, Sharpe, volatility by strategy) sit in figures that are not in the extracted text, so some figures cannot be checked here (inferred).

Items marked (inferred) are our reading of the design, not statements by the authors.

Why it matters

The paper is a widely cited reference for a rule-based, low-turnover trend overlay across asset classes. It shows that the main benefit reported is drawdown and volatility reduction, not higher returns in every year. Practitioners can use it as a baseline when comparing tactical allocation designs.

Questions about this paper

What is the trading rule in Faber's paper?

Buy when the monthly price is above the 10-month simple moving average. Sell and move to cash when it is below. The model is updated once a month at month-end.

How much did the rule reduce drawdown in the five-asset portfolio?

Drawdown fell from 46% for buy and hold to less than 10% for the timing portfolio over 1973-2012.

Does the paper include transaction costs?

No. The backtests exclude taxes, commissions and slippage. The paper discusses them separately and notes turnover of almost 70%.

Where can I read the original paper?

The original is at its publisher: https://doi.org/10.3905/jwm.2007.674809. We link to the paper and do not host the PDF.

Follow new quant research

Download MacroFinance market data as CSV or Excel, or pull it by API, and test the ideas in these papers yourself. New summaries are added as we review them.

Create a free account
Summaries are written by MacroFinance from the published papers, in our own words, for information only. They are not investment advice and may omit detail. Papers are not hosted here: each page links to the original at its publisher.