I’ve decided to put together this lengthy thread because it seems like there’s a dearth of information for active movers and market timers here at TSP Center. However, there’s not much information readily available on more passive seasonal strategies since Jahbulon went quiet a while ago when he started his TSP Holy Grail Research Lab and tspmarketwatch.com sites (both of which are no longer up and running, BTW). While I admit to reading, and following, much of his research (and some of it is replicated here), I’ve also incorporated other things I’ve learned from other sources as well.
Please be aware that this thread isn’t an effort to prove that “Seasonal Beats All”, or that it’s impossible to beat the market using timing or that buy-and-hold is for lazy folks. It’s simply an effort to give more options to TSP investors. I’ve seen more than a few new (or newish) members asking about the seasonal calculator or other seasonal topics lately, and I’ve been asked a question or two in private, so I figured that since I’ve done a lot of studying on this method of investing as it relates specifically to the TSP, I’d share what I know. In the end, we all have the same goal – to make money, right? The data and trends I discuss here are convincing enough to me that I’m following a seasonal strategy with my personal and Fantasy TSP accounts. I’m putting my money where my mouth is.
I’ll caveat all of this by saying that I’m not a fiduciary, I’m not an advisor, I’m not an expert, and I’m not responsible for your monetary decisions. This thread is partly to inform, partly to document my own thoughts and musings, and partly to invite discussion on the topic of seasonal investing within the framework of TSP. You’re all adults here (at least by law if not by mental faculty), and you each hit the “submit” button on the Interfund Transfer screen on tsp.gov of your own accord. That said, let's get this show on the road, shall we?
Seasonal investing is a strategy that looks at historical monthly patterns in the market, analyzes the times of year that have good odds of having positive returns, and then playing those odds and letting the Law of Averages win out over the long term. This method of trading is a good option for those that don’t want to “set it and forget it” because they feel that they need to be aware of what’s happening with their money, but don’t have the time to devote to learning how to time the market or sitting in front of candlesticks for 2-3 hours a day. It’s also an option for the followers that like John Bogle’s take on investing (buy and hold philosophy), that the best strategy is to stay in for the long haul and weather the storms, but don’t completely agree with his rationale when it comes to staying in the market when the odds are that it will be a negative month.
Think of it this way: an MIT study on casino gambling found that the best possible odds for games where you play against the house (instead of other players) is Craps. The mathematical calculations show that the best odds (meaning, risk vs payout) you can get are just shy of 50/50 if you bet the Pass Line and take the Maximum Odds option whenever you have a chance. Any other game in the casino is worse for you from there. Knowing that, if your goal is to have the best odds possible at the casino, why go anywhere else but the Craps table?
Seasonal investing is the same way. We’re taking a look at the historical returns, calculating the averages for each fund for each month, and picking the best fund to be in for that month based on the odds. It’s not “Sell in May and go away”, but it certainly follows the concept that there are certain times when it’s better to be in than out, and vice versa. We’re betting on the odds. Market Timers do the same thing, just using different data and on a shorter timeframe to figure out their odds. “Buy and Hold” followers do the same thing in that they rely on the long-term market average to continue being the average for the next 30 years or so simply because it’s been that way for the past 200+ years. We’re all betting the odds, we’re just using different sets of data to determine our odds.
Note: All the average figures I quote will use the Cumulative Annual Return Rate (CAGR). This number is different from a straight mathematical average and takes into effect negative compounding. Imagine you have $1,000 and have a 100% gain. This gives you a total of $2,000. Then you have a -50% loss, which takes you back to $1,000. The straight average would be a 25% gain, but the real number is 0% because it takes into account the compounding factor. CAGR is the 0% figure, and it’s always lower than a straight average if there’s any negative numbers in the formula, so it makes sense for us to use it so we don't overestimate the actual returns. TSP uses CAGR when figuring the multi-year average on your return for the 10-year fund returns on tsp.gov.
An example of how seasonal strategies work is like this: since 1988 the S fund has been positive in December 23 out of 26 times, or 88% of the time. Over that time, it has a CAGR of 3.4%. In the last 20 years it has been positive 17 times (85%), with a CAGR of 3.22%. And in the last 10 years has been positive 8 times, with a CAGR of 2.81%. The last 5 years has been positive 4 of 5 times, and the CAGR has been 3.87%. Of those 3 negative Decembers since 1988, the returns were -4.32% (2002), -0.4% (2007) and -0.04% (2012). An investor using these historical figures would conclude that there’s a darn good chance of December being a positive month for the S Fund, but if it ends up being a negative month then it’s likely to be minor, so his risk of being in the market during December is small for the gains he's likely to get. December is a good bet.
The seasonal strategy takes this line of thinking and applies it to the rest of the year, allocating a “best fund” to each month based on historical odds. Then the strategy is re-analyzed each month to decide if a particular month’s fund needs to be changed, or if it should stay the same for next year. If the investor desires it, additional study can be devoted to learning “why” each month tends to be positive or negative for particular sectors of the market. December, for example, is affected by fund managers needing to pad their annual numbers, end of year dividends and pay bonuses being rolled over into further purchasing to avoid tax burdens, the Holiday shopping season and by general optimism.
It’s not a 100% accurate method, but then again no strategy is. Personally, I’m satisfied with an 80% success rate – if the strategy works 80% of the time to give me a CAGR of 12% or more per year, then I’m winning in the long run. Why 12%? Because the S&P 500 (C Fund) has a historical 1988 – 2013 CAGR of 9.9% while the Wilshire 4500 (S Fund) has a CAGR of 11.86% for the same time period. If I can beat those benchmarks, then I will be winning over Buy and Hold.
Seasonal investing can have elements of both buy-and-hold methodology and technical chart analysis in it. It’s like buy-and-hold in that you follow a system and largely ignore the media, blogs, and world politics when making your investing decisions. It’s like technical analysis in that you can use technicals to decide when the best time is to make your move if you’re not satisfied with just moving funds at the end of the month. Using technicals can give you +/- a few % each year due to moving at the most optimal time +/- a few days. I won't cover technicals in this thread, that's something that's amply covered by other folks on the Forum.
Like any system, though, seasonal systems have Pros and Cons.
- Emotion is removed from the decision making process. Everything is analyzed logically by establishing quantified limitations of what a “good” month looks like, what the odds are of having a positive month vs the likelihood of a month being negative.
- It’s a systematic approach that analyzes the data and makes adjustments as necessary. You don’t blindly follow the system, you understand “why” you’re in a certain fund at a certain time and have factual data to support that understanding.
- It tends to produce consistent returns.
- Simple: only 6 moves a year.
- If you commit to the strategy, you must follow through with it and ride the emotion that comes with rough times. I made the mistake of not doing so in Feb 2014 – the market took a short dive in late Jan while I was in the S fund. I got scared and ran to the F fund for the duration of Feb because the F fund is the second-best fund for that month. While that Feb ’14 was a positive for the F fund (0.62%), the S fund had a return of 5.43%. This mistake on my part has made a big difference in my ’14 real-world return so far. I’ve learned my lesson: follow the data, trust the numbers. The Law of Averages works.
- No system works all of the time. Sometimes it just won’t work. We’re playing the odds with this strategy and while the Law of Averages says we should win in the long run, it also says that we’ll also have periods when we don’t. What we’re doing is stacking the deck in our favor and making an informed decision on where and when we invest. It’s the long-term end result you need to keep your focus on, not the short term dips and peaks. Followers of Bogle’s philosophy can appreciate this sentiment.
- Times when the markets move sideways for a long time are frustrating to this system because we’re looking for clear positives and negatives during certain times. At such times, we have to resist the urge to try to play Market Timer and try to do better. You avoid doing this by remembering that a small positive is still a positive, and a small negative could be much worse if we aren’t good at our market timing decisions.
So what does it take to follow a system like this?
- An interest in paying attention to what your account is doing while acknowledging that we don’t know enough to play Market Timer well enough to produce reliable returns.
- Dedication to follow a logical system without letting emotion get in the way.
- A long-term focus: 12% a year over 30 years turns a biweekly $200 allocation into around $1,500,000. This would represent a FERS employee that makes $52,000 a year allocating 5% and gets the 5% match for 30 years, and doesn’t get a pay raise the entire time.
- A willingness to ignore how well or poorly everybody else is doing, what the media is saying, whether or not the “experts” are saying to buy or sell, etc. Let the numbers do the talking for us and aim to get the average as positive as we can. The major exception to this is paying attention to the policies set out by the Federal Reserve – they will absolutely impact the F Fund, and to a smaller extent the G Fund.
You might be thinking “That’s all fine and dandy, but what data is out there to support a seasonal system?” Answer: tons of it. Books have been written on seasonal trends in general. http://www.tsp.gov
has historical data going back to 1988 for the G, F and C funds, and it’s easy to find similar data for the S fund (which TSP has back to 2001). I’ve taken the time to make an Excel workbook that details all of the monthly returns for each fund going back to 1988 (I fund goes back to Jan 2001) and calculated the CAGRs as well as the annual Positive / Negative Rate (PNR) for each fund for each month. Most months are pretty black and white as to whether or not it’s a good time to be in the market according to the odds. Then I track various seasonal mixes using that data.
What I’d like to do with this thread is invite discussion on this strategy, and to document the monthly progress of several seasonal TSP fund mixes that I’ve seen, including the one I’m using. Going forward I’ll also do a monthly entry that documents how each fund has performed in the past for the next upcoming month, and this entry will be towards the end of the month so interested parties will have enough time to percolate on what the upcoming month looks like historically. I’ll also post how the various mixes are set for the upcoming month.
And with that, I’ll end this lengthy post and use the next few posts to do an overview of the various fund’s historical details and the seasonal mixes I’m tracking. They are as follows:
- TSP Center’s default setting on the “TSP Seasonal Calculator” page.
- Jahbulon’s Basic Seasonal formula
- Boltman’s Seasonal Formula
- A straight “Sell in May and Go Away” using G and C
There’s other mixes I’m tracking, but none are doing well. I can give those stats as well if anybody is interested and asks nicely. Also, if anybody has another version they’d like me to add to the list and my ongoing tracking database, let me know and we’ll see if it’s a promising mix. None of the mixes I follow use the L funds, and I don’t anticipate starting them anytime soon. The L funds don’t fit into the methodology and mindset that seasonality uses.
Stay tuned for the individual Fund History posts.