Is the IHSG Falling Victim to the “SeptemBear” Phenomenon?

Over the past ten years (2016-2025), the Jakarta Composite Index, aka IHSG, has closed September in negative territory seven times, and positive only three times. Roughly speaking, statistically, the probability of the IHSG weakening in September reaches 70%.

Interestingly, September isn’t the only month with a “reputation” among investors. There’s also “Sell in May”, which flags May as a correction-prone month, and “Window Dressing”, which makes December a favorite among many.

These three phenomena are often grouped together as calendar anomalies, recurring market patterns driven not by economic fundamentals, but by the habitual behavior of market participants at certain times of the year. The question is: are all three equally valid?

Why Does September Fall Into Such a Bad Pattern?

Picture the habit many people have after returning from a long holiday, say, after the Lebaran homecoming or a year-end break. Once back to routine, the first thing people usually do isn’t diving straight back into work, but rather “tidying up” first, re-evaluating priorities that had been put on hold.

A similar phenomenon happens in the global finance industry, on a much larger scale. Throughout the summer (June-August), many fund managers are on vacation, so trading volume tends to thin out. Once September arrives and they return to active duty, portfolio rebalancing occurs: many major institutions evaluate and restructure their portfolios simultaneously within a short window, creating significant selling pressure on the market.

The IHSG Data of SeptemBear

IHSG Performance in September (2016–2025)

Over the past 10 years (2016-2025), seven years closed negative and only three closed green (2017, 2021, 2025), with an average return of -0.88%. The deepest decline occurred in September 2020 (-7.03%), coinciding with the sudden announcement of strict social distancing measures in Jakarta due to a surge in Covid-19 cases, which triggered panic selling to the point that trading was temporarily halted.

But the three anomalous years above actually reveal something important: this pattern can be overridden by strong fundamental sentiment. September 2017 was green because Bank Indonesia unexpectedly cut its benchmark interest rate. September 2021 was green thanks to a commodity rally that pushed Indonesia’s export value to a record high. September 2025 turned green again, supported by major banking issuers’ earnings coming in above market expectations along with a stable Rupiah. In other words, strong domestic sentiment can help sustain the market’s upward trend, even when the seasonal pattern leans toward weakness.

If September Marks the End of Q3, Why Doesn’t It Follow the Window Dressing Pattern?

This is the interesting part. Window Dressing is actually not just a December phenomenon. By definition, it occurs at the end of every quarter (March, June, September, December), since investment managers report portfolio performance every three months. That means September, which also marks the close of Q3, should theoretically get its share of the positive “portfolio tidy-up effect”.

But in reality, September has been dominated by selling pressure over the past decade. This shows that global rebalancing pressure and negative seasonal sentiment in September far outweigh the relatively weak quarterly window dressing push. It’s only in December that window dressing truly makes its impact felt on the IHSG, because that’s when there’s additional pressure: annual performance reports, not just quarterly ones, making the institutional push to “dress up” portfolios far stronger.

The data backs this up: over the past 10 years, December recorded 8 green years with an average return of +2.46%, far more consistent than September. The two anomalies (2022 and 2024) show that even window dressing can “fail” when selling pressure on major sectors is too strong, or when foreign outflows continue due to global uncertainty.

So what about Sell in May? May’s own data (2016-2025) shows 7 out of 10 years negative, with an average of -0.67%, a level fairly similar to the SeptemBear Effect, even though globally this phenomenon is more commonly known as “weak May–October, strong November–April.” In the IHSG, May’s negative pattern is fairly consistent, but it isn’t necessarily followed by continued weakness in the following months as the original theory claims.

Average Returns for Three Capital Market Calendar Phenomena (2016–2025)

Why Can an Effect From the Global Fiscal Calendar Spread to the Domestic Market?

The answer lies in the interconnectedness of global capital flows. Foreign investors hold a significant share of ownership in large-cap stocks on the Indonesia Stock Exchange (IDX). When global institutions rebalance their portfolios, their decision to reduce exposure to emerging markets, including Indonesia, also triggers selling activity on the domestic exchange. In addition, September often coincides with the Fed’s interest rate decision period, whose outcome also influences Bank Indonesia’s policy direction and the Rupiah’s movement.

Of these three calendar phenomena, the SeptemBear Effect and Sell in May have both proven fairly consistent based on the IHSG’s historical data, while December’s Window Dressing stands out as the most reliable of the three. However, this seasonal pattern can still be completely overridden by stronger fundamental sentiment, whether from monetary policy, commodity prices, or issuer earnings performance.

---

Disclaimer: This content is created for educational purposes or service promotion, and does not constitute a recommendation to buy or sell any specific Securities. Any risks arising from investment decisions made based on the information in this publication are the sole responsibility of the respective audience. The data was sourced from various sources. PT KAF Sekuritas Indonesia is licensed and supervised by the Financial Services Authority (Otoritas Jasa Keuangan / OJK).