BUY AND HOLD?

The S&P 500 flirted with a "supposedly" important technical level – the 20-day moving average – on Friday. I decided to back-test the historical (May 19, 1982 to February 20, 2009) performance of two simple rules - trend-following and contrarian - based on this technical parameter.

Rule #1: Trend-following
Buy on the close, if S&P 500 > 20-day moving average
Liquidate on the close, if S&P 500 < 20-day moving average

Rule #2: Contrarian

Buy on the close, if S&P 500 < 20-day moving average
Liquidate on the close, if S&P 500 > 20-day moving average

To say the least, the results for the simple trend-following strategy are quite dismal. The order of return dominance (See Exhibit 1) over the entire back-testing period is the following:

1. Buy & hold: $1 invested in the S&P on May 19, 1982 would have grown to $6.71 by February 20, 2009 (7.37% average annual return).
2. Contrarian: $1 invested in the S&P on May 19, 1982 would have grown to $3.35 by February 20, 2009 (3.35% average annual return).
3. Trend-following: $1 invested in the S&P on May 19, 1982 would have grown to $2.00 by February 20, 2009 (2.63% average annual return).

However, there are important variations over different time periods. The trend-following strategy broadly outperformed until about 1998. Since then the contrarian rule held sway. Buy & hold proved the best across time intervals.

Exhibit 1: Value of $1 Invested in Different Strategies: May 19, 1982 to February 20, 2009



Exhibit 2: Relative Performance (Trend Following/Contrarian): May 19, 1982 to February 20, 2009



Source: Bloomberg