Between February and May 2026, the DSE banking sector lost Tk 133.9 billion in market capitalisation. Over roughly the same window, pharma added Tk 32.3 billion. The DSEX moved less than 50 points over that period. If you watched the index alone, you saw nothing. If you watched the sectors, you saw money quietly walking out of one room and into another.
That movement has a name. It is called sector rotation, and on a market as concentrated as the DSE — where banking alone represents the single largest sector by market cap — it is the most important signal most retail investors never learn to read.
The good news is that the data needed to read it is already public. The bad news is that no Bangladesh market resource explains how. So here is the framework.
Signal One: Market Cap Change Tells You Where Money Went
The first place to look is not the index. It is the sector-level market capitalisation change over rolling 30, 60, and 90-day windows.
Banking ran Tk 806.9 billion in February. It closed May at Tk 673.0 billion. That is a 16.6% contraction — and crucially, it happened during a period when the overall DSEX barely moved. Money did not leave the market. It left the sector. Pharma rose 9.0% over an overlapping window, from Tk 359.2 billion to Tk 391.5 billion. Engineering crept up from Tk 258.1 billion to Tk 262.7 billion and continued climbing.
The arithmetic alone is the rotation. Capital does not vanish; it relocates.
What confirms relocation rather than panic is what happens at the index level. When DSEX falls in lockstep with sector market caps, you have a market-wide sell-off. When DSEX holds while one sector bleeds and another swells, you have a rotation. The May data is the second case.
Signal Two: Turnover Spikes Mark the Handoff
Sector rotation almost always shows up in turnover before it shows up cleanly in prices. The day institutional money repositions, it leaves a fingerprint in volume.
On 14 May, DSE turnover crossed Tk 1,101 crore — up 54% from the prior session and the first crossing of the Tk 1,000-crore mark in weeks. That spike was not random. It coincided with the DSEX snapping a five-day losing streak. High turnover at the bottom of a slide, followed by a recovery led by non-banking names, is the textbook handoff signal. Selling concentrated in outgoing sectors meets buying that initiates the incoming ones.
A general rule: when daily turnover jumps 30% or more during or just after a sell-off, do not assume capitulation. Look at which sectors absorbed the volume. That is where the next leg is forming.
Signal Three: Breadth Tells You Whether It Is Rotation or Panic
The same 14 May session had 194 decliners against 161 advancers — negative breadth on a day the index recovered. That combination is almost never random.
In a panic, breadth and index move together — both decisively down. In a rotation, breadth stays negative even while the index rises, because the declines are concentrated in the sector being exited and the advances are concentrated in fewer, larger names being accumulated. The 165-to-180 decline ratios seen earlier in May were the leading edge of the same phenomenon.
Read breadth and the index together. They tell different stories. The rotation story is when they disagree.
Signal Four: The Valuation Gap Explains the Why
Rotation without a reason is just noise. The reason is almost always relative valuation against earnings growth.
Banking trades at a PE of 6.8x against a 3-year average of 7.3x — already cheap, but earnings growth is just 3.5% annually. Pharma trades at 9.8x against a 3-year average of 12.8x — meaning the sector is priced below its own historical norm despite earnings growing at 8.0%. A buyer comparing the two sees pharma offering more than twice the earnings growth at a relative discount. The BXPHARMA rally on 5 May and the broader pharma gainers list — Asiatic Laboratories up 316.9% over one year — are the consequence, not the cause.
The cause is the math. Cheaper growth attracts capital. Period.
Putting the Framework Together
The four signals work as a single instrument. Sector market cap change tells you where money went. Turnover tells you when the handoff happened. Breadth confirms whether the move is rotation or genuine panic. Valuation against earnings growth tells you why.
When the May 2026 data is read together — banking down 16.6%, pharma up 9.0%, turnover spiking 54% on 14 May, negative breadth during an index recovery, banking earnings growth of 3.5% versus pharma’s 8.0% — you have the complete fingerprint of an institutional rotation.
The DSEX closed 24 May at 5,336, up 13.06% year-on-year. The investors who learned to read the rotation captured most of that move from inside the right sectors. The investors who watched the index alone saw a sideways market.
That is the difference the framework makes.