Tis’ the Season to ‘Get Small(Cap)’

  • SumoMe

A lot has been written lately about the tendency for small-cap stocks to outperform large-cap stocks during the upcoming time of the year (for example).  And rightly so.  For whatever reason, the relationship between large-caps and small-caps has shown itself to be highly seasonal over the years. Let’s take a closer look.

Jay’s Large-Cap/Small-Cap Calendar

The table below displays my calendar for when to hold large-cap stocks versus when to hold small-cap stocks.

*Note that TDM stands for “Trading Day of the Month” and NOT date of the month – i.e., Mar 10 means at the close on the 10th trading day of March small-caps are sold and large-caps are bought.

*Also, for June and September if there are less than 22 trading days in the month then the trade is made at the close on the last trading day of the month.

From Close on Trading Day of Month To Close on Trading Day of Month Hold
Mar 10 May 12 Large-Cap
May 12 Jun 22 Small-Cap
Jun 22 Aug 6 Large-Cap
Aug 6 Sep 22 Small-Cap
Sep 22 Nov 15 Large-Cap
Nov 15 Mar 10 Small-Cap

Figure 1 – Jay’s Large-Cap/Small-Cap Calendar

The Test

*For testing purposes we will use ticker RUI (Russell 1000 Large-Cap Index) and ticker RUT (Russell 2000 Small-Cap Index).

*First we will look at a buy-and-hold approach using both indexes, then will compare it to our switching method.

The Results

Figure 2 shows the growth of $1,000 invested in each index separately as well as combined since January 4, 1989.2Figure 2 – Growth of $1,000 invested in RUT and RUI (separately and “Split” between the two); 1/4/1989-11/21/2017

Figure 3 applies the calendar displayed in Figure 1 –  i.e., Ticker RUI is held from March TDM 10 through May TDM 12, the last trading day of June through August TDM 6 and the last trading day of September through November TDM 15 (with all switches being made at the close of the day). Ticker RUT is held during all other periods as indicated in Figure 1.

Note: The “red” line in Figure 3 is the “green” line from Figure 2 and represents splitting an initial $1,000 investment between RUI and RUT on a buy-and-hold basis.

Figure 3 then displays the “Switching” results versus the “Splitting” results (i.e., simply buying and hold both indexes).

3Figure 3 – $1,000 invested using “Switching” strategy versus “Splitting” money evenly between two indexes; 1/4/1989-11/20/2017

For the record, the “Switching” strategy gained +3,727% since 1989 versus +892% for the “Splitting” strategy.  While the Switching strategy return is 4.17 times greater than simply splitting money between both indexes, it does not come without risk and volatility. To wit:

*The average 12-month standard deviation of returns is actually higher (18.1%) using the Switching strategy versus splitting (16.8%)

*The Worst 12-month % decline for the Switching Strategy is still a whopping -45.8% – almost as much as the -49.0% for splitting.

Measure Switch Split
Average 12-mos. % +14.5% +9.2%
Median 12-mos % +14.8% +10.9%
Std. Deviation% 18.1% 16.8%
Average/Standard Deviation 0.80 0.55
Worst 12.mos. % (-45.8%) (-49.0%)
% 12-mos. UP 82.1% 76.3%
% of 12-month periods outperforms 84.1% 15.9%

Figure 4 – Comparative Results: “Switching” versus “Splitting”; 1/4/1989-11/20/2017

Figure 5 displays the year-by-year results.

Year Switch Split Diff $1,000 $1,000
1989 22.9 19.4 3.5 1,229 1,194
1990 (9.5) (14.1) 4.6 1,112 1,025
1991 40.7 35.3 5.4 1,565 1,387
1992 17.6 10.7 6.9 1,841 1,536
1993 13.1 12.0 1.0 2,082 1,721
1994 4.6 (2.8) 7.4 2,177 1,672
1995 31.0 30.3 0.7 2,852 2,179
1996 21.0 17.3 3.7 3,450 2,555
1997 32.6 25.7 6.9 4,577 3,213
1998 11.8 12.0 (0.2) 5,116 3,599
1999 14.8 18.5 (3.7) 5,873 4,266
2000 28.1 (6.2) 34.3 7,522 4,000
2001 3.8 (7.6) 11.4 7,807 3,695
2002 (17.0) (22.3) 5.4 6,483 2,870
2003 35.2 35.6 (0.3) 8,768 3,890
2004 27.2 13.1 14.1 11,157 4,401
2005 6.6 3.8 2.8 11,898 4,570
2006 18.1 15.2 3.0 14,053 5,263
2007 5.4 0.5 4.8 14,807 5,290
2008 (30.8) (37.0) 6.2 10,247 3,335
2009 24.6 25.3 (0.8) 12,765 4,180
2010 20.5 19.6 0.8 15,378 5,001
2011 (2.1) (3.1) 1.0 15,058 4,845
2012 21.0 14.2 6.8 18,220 5,535
2013 40.7 33.9 6.8 25,642 7,411
2014 14.8 7.1 7.7 29,437 7,936
2015 1.5 (3.5) 4.9 29,875 7,661
2016 12.0 14.6 (2.5) 33,475 8,779
2017 14.3 13.0 1.4 38,273 9,917

Figure 5 – Year-by-Year Results; 1/4/1989-11/20/2017

The next switch occurs at the close on 11/21 (15th trading day of November 2017 out large-caps and into small-caps).

Summary

The “system” (such as it is) described herein is by no means of the “you can’t lose” variety (note the “Worst 12-month % loss” of -45.8%).  Still, the fact that this entirely mechanical – and calendar-based – approach outgained a simple buy-and-hold approach by over 4-to-1 overall, and during 84% of all rolling 12-month periods – is nothing to sneeze at.

Jay Kaeppel

Disclaimer:  The data presented herein were obtained from various third-party sources.  While I believe the data to be reliable, no representation is made as to, and no responsibility, warranty or liability is accepted for the accuracy or completeness of such information.  The information, opinions and ideas expressed herein are for informational and educational purposes only and do not constitute and should not be construed as investment advice, an advertisement or offering of investment advisory services, or an offer to sell or a solicitation to buy any security.

 

 

 

6 thoughts on “Tis’ the Season to ‘Get Small(Cap)’

  1. Tested it with TF and ES futures contracts and looks great. 25k draw using a single contract. Hard system to go off and turn on given the run the market has had.
    Jay Given you have so many seasonal edges how to you go off an allocate across them?

    Cheers
    Ben

  2. So per your research we would be at an advantage holding EEM today, yet we should be out of the market given today is a -5TDM day. How do you handle these conflicting “signals”?

    1. 3500k on TF and 10k on ES doesn’t really effect the system. draw is now 20k
      Because the stockmarket has mean reverting tendency it’s difficult to place stop without affecting switches.

  3. 3.5k on TF and 10k on ES doesn’t really effect the system. draw is now 20k
    Because the stockmarket has mean reverting tendency it’s difficult to place stop without affecting switches.

    1. The next system [The 40-Week Cycle (Begin Again)], uses a 12% stop loss from entry. Do you think that system avoids mean reversion? It seems to me that some sort of stop loss would prevent that 40+% draw down this system reports.

      I like this system a lot, but I could not take the draw down. So, I would need to set some sort of stop beforehand to make it work for me.

Leave a Reply

Your email address will not be published. Required fields are marked *

This blog is kept spam free by WP-SpamFree.