Trend Following 1900-1949

Summary

The last 12 months have been a particularly hard time for Tines Capital’s active strategies. When strategies of any type are performing poorly, it makes sense to get nervous. Of course, patience and consistency are often the appropriate choice. One of my personal antidotes to getting nervous is diving into what has happened in the past. Though past performance doesn’t guarantee future results, I still find history important and helpful. Though history doesn’t repeat itself, it often rhymes.

Today, I want to take a look at a simple trend following strategy and how it might have performed from the period of 1900-1949 which includes WW1, Spanish Flu, The Great Depression, WWII, and many other important events most of us don’t remember learning about.

The Data

From 1900 through 1949 end of month data is available for SPX (S&P 500) from tradingview.com. In the chart below you can see the performance of SPX. As you can see there are some really good and bad periods. Dividends were a larger part of the returns in this period of history, unfortunately tradingview.com does not provide that data. If someone had just taken a passive approach with dividend reinvestment the performance would have been better than indicated in the tradingview chart below.

SPX from 01/01/1900 to 12/31/1950

Simple Strategy Rules

This strategy is not one that Tines Capital uses directly and this simulation will be simplified for the sake of the post.

Buy or hold SPX if previous month has a positive return.

Sell SPX or hold cash if previous month has a negative return.

The price of SPX was available for most months. Occasionally, a month isn’t available in the data. When that occurs the closest available month was used.

Results

Analysis

Over the included period this rudimentary trend following strategy did very well. This doesn’t mean it always would, and as mentioned below there are many important caveats. However, I think this exercise is helpful in seeing why trend following can at times be helpful. Though there are periods of whipsaw and bad performance, I believe trend following should be a part of most client’s investment portfolios.

You can view the calculations here:

Important Caveats

  • The effect of dividends isn’t included for the strategy or SPX. As shown in this image obtained here, dividends were quite large during this period. However, since neither the strategy or SPX includes the dividends in the simulation it should not completely invalidate the lesson.
  • The effect of taxes and fees are not included. These can be significant and should be included when deciding on investment methods.
  • Just because a certain method shows promise in backtesting, we cannot assume it will do well in the future.
  • In many ways this period was a golden age of trend following. There are other periods where trend following doesn’t do as well.
  • As always, I welcome an alert on any typos or errors you notice.