[Editor's Note: This is an old article, originally published in 2002, but we keep it available as an introduction to the annual seasonality concept. Much has changed since then at SMI however. We now believe that our newer Dynamic Asset Allocation strategy provides better signals for those wishing to alter their exposure to the stock market on an ongoing basis.

In our ongoing search for easy-to-use mechanical strategies, we recently came across this method of improving the historical returns of Just-the-Basics from 12.3% to 16.8% per year (from 1988 through 2001) while at the same time cutting its risk by half. Too good to be true? Not likely, as other studies have shown similar results going back over half a century.

SMI's Just-the-Basics indexing strategy claims to be the ultimate in low-maintenance investing. Once per year, you balance your portfolio between the recommended funds, then walk away for twelve months and let your portfolio run on autopilot. Your reward for such simplicity is a lot of time not spent thinking about investing throughout the year, and a rate of return roughly equal to the overall stock market. But what if there were an easy way to get the same returns — or better — while cutting your risk in half, all while maintaining the same simple, hands-off approach that likely attracted you to Just-the-Basics in the first place? Believe it or not, there is, and with only a slight modification to Just-the-Basics as it is normally implemented.

Let's quickly identify the target audience of this article. Readers who follow SMI's portfolio recommendations typically divide into two camps. In the first camp are those investors whose goal is to beat the market's rate of return, and they're willing to be actively involved with their investments each month in order to do it. For these investors, we offer the Fund Upgrading strategy. While Upgrading may not beat the market's return every year, over time it offers the highest potential returns of any SMI investing strategy, although it requires the most effort as well. If you're an Upgrader, chances are this article will be interesting but not something you want to switch to unless you're specifically wanting to reduce your time involvement.

In camp number two reside those readers who either don't want to be involved with their investments as often as Upgrading requires, or who remain unconvinced that the added effort will produce improved returns. These are our Just-the-Basics investors, who are willing to accept results that are roughly in line with the overall market in return for an easier, simpler strategy. It is to these readers that this article is primarily targeted.

Seasonality explained

There's an old stock market saying: "Sell in May and go away." The idea is that historically a lot of Wall Street types have tended to lighten up on their stock holdings over the summer, presumably as they went away on vacation. But until fairly recently, this was more of a "folk wisdom" type of thing. Experienced market watchers had recognized, but not really quantified, the fact that volume is lighter and stock returns weaker over the summer than during the winter months. With the arrival of affordable computers in the 1980s, it was just a matter of time before someone undertook the number crunching necessary to learn whether this seasonal tendency was significant.

One groundbreaker in this area was Yale Hirsch, publisher of the Stock Trader's Almanac. Hirsch discovered in 1986 that the market tends to have two distinct phases each year: a "favorable period" lasting from November-April, and an "unfavorable period" running from May-October. Obviously, this pattern doesn't hold true every year, but over the past half century the trend has been unmistakable.

In the latest update of his research (Stock Trader's Almanac 2002, page 52), he shows that an investor who had invested in the Dow Jones Industrials during just the favorable six months of each year from 1950-2000 would have earned a whopping profit of $425,890 on an initial investment of $10,000. By contrast, investing the same $10,000 during only the unfavorable six months of each of those years would have earned a mere $1,743! The total time invested in the market would have been exactly the same, but the returns are wildly different. The implication is clear — most of the stock market's gains over the past half-century were attributable to the six-month favorable period beginning in November and lasting through April.

What causes the annual seasonality effect?

With any good mechanical system, it's important that there be a credible rationale as to why it works. Is there a good explanation as to why these distinct favorable and unfavorable periods occur? Absolutely. Consider the flow of money into investors' hands. Much of it arrives in the form of year-end performance and Christmas bonuses, distributions to owners from profitable small businesses that close their books in January and February, income tax refunds in the first few months of the year, mutual fund distributions in December, and other "lump-sum events" that tend to recur during the favorable period. As these extra sums of money arrive, much of it makes its way into the stock market.

Now consider what happens during the unfavorable period from May through October. Beginning in April, those not receiving income tax refunds actually end up having to pay taxes, sometimes requiring them to sell stocks to obtain cash. Money is set aside for summer vacations, and then those vacations actually occur, causing investors to pay less attention to the market. In the case of large investors, they may lighten up on their stock holdings before leaving for vacation, lest misfortune strike while they are away.

The general trend is clear: during the favorable months, there are multiple factors at work putting new money into the market, creating healthy buying which drives the market up. During the unfavorable months, at best there is less new money propelling the market forward, and at worst there are factors working together to cause money to be pulled out of the market. As the inflows of new money dry up, the market tends to drift sideways, or even decline.

Combining annual seasonality with monthly seasonality

In his book, Riding the Bear, newsletter publisher Sy Harding provides a valuable addition to this annual seasonality strategy. As Austin has pointed out before, it has long been recognized that there is also a monthly seasonality pattern at work in the stock market. This is separate from the annual seasonality we've been discussing so far. Research has shown there is a favorable six- to seven-day period surrounding the change of each calendar month that has historically provided vastly superior returns than those generated during all the rest of the trading days in the month put together.

Harding combines this knowledge of monthly seasonality with the broader annual seasonality Hirsch discovered. So, rather than buying November 1, Harding suggests buying earlier, just ahead of the October month-end favorable period. And rather than selling on the last trading day of April, he recommends waiting and selling a few days into May at the end of the favorable period that is in progress. Simple enough — we're just buying stocks a little earlier and selling them a little later. Other than that, it's the same basic "six months in, six months out" strategy.

The results from this simple adjustment are stunning. Harding studied the results of investing in the Dow for just the six-month favorable period, plus the few days on either end, as we just discussed. The rest of the year, he assumed the money was invested in short-term T-bills. Over the 35-year period from 1964-1998, Harding's strategy would have caused a $10,000 investment to grow to $787,200. What would you have earned if you'd invested the same amount and left it fully invested year-round during the entire thirty-five year period? Just $425,200. Despite being invested in stocks just half the time, returns were almost double that of a strict buy-and-hold approach, while risk was cut nearly in half, as six months of each year were spent completely out of the stock market!

Applying seasonality to SMI's Just-the-Basics strategy

How can we apply these findings to improve the performance of SMI's model portfolios? As alluded to earlier, the best application is probably not with our Fund Upgrading strategy. The reason for this is that Upgrading seeks to find managers with styles that are performing well at that given point in time. Often times, a good manager can do quite well even when the broad market is flat to lower. So, our initial impression is that limiting our Upgrading efforts to just half the year doesn't seem to make sense.

However, given the impressive results and the simplicity with which this seasonality approach can be executed with indexes, applying it to Just-the-Basics seems a natural fit. Since JtB seeks to track the market, and Harding's research indicates this can help us beat the market with less risk, it's certainly worth looking at as a potential variation on "regular" JtB. Since JtB is a purely mechanical strategy itself, we decided to back-test this new approach as far into the past as we could, which turned out to be the beginning of 1988 when the Vanguard Extended Market Index Fund (one of JtB's core holdings) was introduced.

First, we had to decide which version of monthly seasonality to use. As Austin wrote last month, there are two competing ideas as to when this end-of-month rally tends to occur. Norman Fosback, the originator of the monthly seasonality idea, defines it as the last two trading days and first five trading days of each month. Yale Hirsch though, believes that as more investors have become aware of this trend, the rally itself has shifted earlier. He believes the monthly favorable period now begins with the last four trading days of the current month and runs through the first two trading days of the following month.

Table A

After examining the data, we came to the same conclusion as Hirsch. Therefore, for our testing, we assumed you would enter the market at the close of the fifth trading day prior to the end of October, and exit at the close of the second trading day of the month in May. The idea is to catch the beginning of the monthly rally during the last four days of October, then hold for the six-month favorable period, then also catch the end of the monthly rally during the first two days of May.

Table A shows the year-by-year results and how they compare to regular JtB. The exact entry and exit dates, and the profits that resulted from following the strategy using those dates, are shown in Table B. Bear in mind that these returns reflect multi-year averages; applying the seasonality strategy in any one year may not give satisfactory results.

With the buy and sell dates established, the next step is knowing what to buy and sell. This part is easy — we simply used the standard Just-the-Basics allocation for an all-stock portfolio that appears at the top of the Recommended Funds page. Note that in a portfolio with an 80% stock allocation, only 80% of the portfolio is moved into stocks during the favorable period; 20% remains in bonds year-round. Other portfolios with less than a 100% stock allocation are treated Similarly.

During the favorable period, we assumed the money was split 40/40/20 between Vanguard's S&P 500 Index fund, the Extended Market Index fund, and the International Growth fund. Then, entering the unfavorable period, the entire balance was switched over to the Vanguard Total Bond Market Index fund. Each year upon re-entering at the end of October, we re-allocated back to the standard 40/40/20 recommendation.

We saw strong hypothetical results, with this seasonal variation of JtB averaging 16.8% versus 12.3% for regular JtB which was left fully invested year-round for the full 14-year period. Perhaps the results would appear more striking if I gave them to you in dollar terms — it's the difference between $25,000 growing to $218,690 versus $126,840. And remember, this seasonality adjustment also reduced the overall risk by nearly 50%, as half the time during the period was spent out of the stock market in the relative safety of the Bond Index fund.

Would Enhanced Just-the-Basics, SMI's other variation on Just-the-Basics, also be helped by adjusting for these seasonality patterns? We didn't test it, so we can't say for sure. But it stands to reason that at least for the index funds there should be an improvement seen. After all, index funds used are the same ones we tested with here. The Rotation funds are a somewhat different animal, in that we expect them to be cycling through the sectors in the market that are currently in favor, and as a result, it's a more difficult prediction to make as to whether we'd be better served sitting out of them for six months each year. Regardless of whether you hold the Rotation funds year round or just during the favorable period each year, you could always opt to follow the seasonal approach for your index funds.


Much of the appeal of Just-the-Basics lies in its simplicity, so it's important to understand that this variation really doesn't complicate matters very much. While this is easy to lose sight of when reading about our testing, here are the simple steps required to apply this seasonal variation of JtB:

  1. Buy a calendar.
  2. Count back five trading days from the beginning of November. This year, November 1 is a Friday, so counting back five trading days (skipping the weekend) puts you on Friday, October 25. This is the day you would want to purchase your JtB funds. Use the allocations shown at the top of the Recommended Funds page for your individual stock/bond mix.
  3. Forget about the market. Enjoy the holidays.
  4. Sell your stock funds on the second trading day in May. For 2003, this will be Friday, May 2. Transfer all your money into the Vanguard Total Bond Market Index fund.
  5. Forget about the market. Enjoy the summer.
  6. Repeat steps 1 through 5, and, assuming the next decade reflects the pattern of the past half century, watch your returns handily beat the market with half the risk!