Much as September is dreaded by many students as it’s time for back-to-school, this month is also not well-liked by many equity investors with an eye for seasonal patterns as, much like its start this year, September has historically been a poor month for equity markets. Markets traded down (S&P 500 shed over 2%) on the first trading day of September yesterday with the worst daily performance since the start of August, as concerns about global growth came to the fore among market participants. Today, we examine in more detail how September has typically fared from a seasonality perspective.  

Since 1950, the S&P 500 has generated an average return of -0.7% in September, making it the worst month for stocks on an average return basis. The last four Septembers have also been notably weak, with the index posting declines of 4.9%, 9.3%, 4.8%, and 3.9%, contributing to an average decline of -4.2% over the past five-year period. September has the worst monthly return over this period, and also over the past 10- and 20-year periods. It is also the only month that exhibits negative returns over each of these periods. 

September Seasonals Have Been a Headwind Especially in Recent Years

S&P 500 Index Average Monthly Returns (1950–2024)

Source: LPL Research, FactSet, Bloomberg 8/30/24  (1950–current)
Disclosures: All indexes are unmanaged and cannot be invested into directly. Past performance is no guarantee of future results. The modern design of the S&P 500 Index was first launched in 1957. Performance before then incorporates the performance of its predecessor index, the S&P 90.

While this hasn’t been the pattern at the start of September this year, on average since 1950 during the month, stocks tend to trade sideways during the first half of the month, with losses beginning to accumulate towards month’s end. For this year, the midway point also happens to line up closely with the September 18 Federal Open Market Committee Meeting (FOMC) which could be a catalyst for increased volatility, especially if the Federal Reserve (Fed) decision doesn’t meet market expectations (currently futures markets price the probability of a 25 basis point (0.25%) cut at this meeting at around 60%, and a larger 50 basis point (0.25%) cut at around 40%).  We examined “What Does a Rate Cut Mean for Stocks and Bonds” and why we think the markets may be pricing in too many rate cuts in our blogs last week. 

Poor September Performance Has Typically Been Backloaded 

S&P 500 September Daily Progression (1950–2023)

Source: LPL Research, FactSet, Bloomberg 8/30/24  (1950– current)
Disclosures: All indexes are unmanaged and cannot be invested into directly. Past performance is no guarantee of future results. The modern design of the S&P 500 Index was first launched in 1957. Performance before then incorporates the performance of its predecessor index, the S&P 90.

Analyzing the data by the proportion of positive monthly returns doesn’t improve the seasonal outlook for September, as it is the month with the lowest proportion, by some distance, of positive monthly returns. September is the only month since 1950, that has finished positive less than half of the time. 

September Has a Considerably Lower Proportion of Positive Monthly Returns

S&P 500 Index Percentage of Positive Monthly Returns (1950–current)

Source: LPL Research, FactSet, Bloomberg 8/30/24 (1950–current)
Disclosures: All indexes are unmanaged and cannot be invested into directly. Past performance is no guarantee of future results. The modern design of the S&P 500 Index was first launched in 1957. Performance before then incorporates the performance of its predecessor index, the S&P 90.

Before we send September straight to the metaphorical principals office for poor performance, there are a couple of potentially redeeming features for September 2024. In years when stocks have been up year to date by the end of August (50 of the last 73 years, and as they have been this year), then September has actually been marginally positive (0.1%) on average. September in years where markets have been down year to date follows that trend and exhibits an average loss of 2.8%. Additionally, in presidential election years, September’s average returns have been slightly better, losing 0.4% compared to a loss double that size in non-presidential election years. The percentage of positive monthly equity returns in September also slightly improve to around a coin-toss (50%) in both presidential election years and years with positive year-to-date returns.  

Considering seasonality past September, the outlook does improve considerably. The two-month period from October to November is the strongest over the past five and ten years, and second-strongest over the past 20 years, and all periods back to 1950. November is the strongest month based on data to 1950, with December second; as it’s no surprise that November-December is the strongest two-month period over that time frame. The fourth quarter is also the strongest quarter for stock market returns over all the periods studied (including an impressive 9.8% average quarterly gain over the past five years). In presidential election years, only the fourth quarter trails the second quarter (+2.8%) as the strongest, but still posts an impressive 2.5% average gain. A caveat to the improved seasonality past September is that in presidential election years, stocks have also been down in October (-0.9%) as uncertainty over upcoming election results has often spooked markets (compared with October in other non-presidential election years with an average +1.5% return). Presidential election year progression of returns sees stocks peak around the third week in August then bottoming around a week before the election (losing 2.4% on average in the process), before ending the year strong after the election is in the rear-view mirror. 

Stocks Have on Average Bottom Around a Week Before Presidential Elections‍‍‍‍‍‍

S&P 500 Index Daily Progression (1950–2024)

Source: LPL Research, FactSet, Bloomberg 8/30/24 (1950– current)
First 250 trading days of the year presented; actual number of trading days has ranged from 250 to 254.
Disclosures: All indexes are unmanaged and cannot be invested into directly. Past performance is no guarantee of future results. The modern design of the S&P 500 Index was first launched in 1957. Performance before then incorporates the performance of its predecessor index, the S&P 90.

Summary 

September seasonality is generally not favorable to stocks, with poor performance typically backloaded, but there may be some slightly mitigating factors at play in the year. Much of the likely trajectory for stocks will depend on the path that the Fed decides to take at the FOMC meeting that takes place right at the start of the historically weak period for stocks in September. LPL’s Strategic and Tactical Asset Allocation Committee (STAAC) maintains its tactical neutral stance on equities, though we don’t rule out the possibility of short-term weakness as markets could readjust having gotten slightly over their skis on the pace and number of rate cuts that the Fed will deliver. Scanning out further, both post-election and fourth quarter seasonality become more favorable stocks.