Today was a very good example of why you cannot read the 3:30/official close exactly the same way as a normal continuous-market move anymore, especially on an MSCI rebalance day.
Nifty 50 closed at 24,080.40, down 95.25 points or 0.39%. But before the closing auction it was down roughly 0.52%. So CAS actually recovered a small part of Nifty’s decline rather than causing the benchmark sell-off. The underlying weakness during the day came from Brent moving above $90, renewed US–Iran tensions, higher Fed-rate expectations, foreign selling and weakness in names such as HDFC Bank. (Reuters)
The crazy part was what happened inside individual constituents during CAS.
1. MSCI funds had to rebalance today
The new MSCI weights become effective September 1, which means passive funds tracking MSCI had to finish their trades today so that tomorrow morning their portfolios match the new index.
Among the major changes:
| Stock | MSCI action | Mechanical flow |
|---|---|---|
| Laurus Labs | Added | Large buying |
| Adani Energy Solutions | Added | Buying |
| Lenskart | Added | Buying |
| Groww | Added | Buying |
| Eternal | Weight increased | Buying |
| Adani Enterprises | Weight increased | Buying |
| Reliance | Weight reduced | Selling |
| SBI Cards | Deleted | Selling |
| Astral | Deleted | Selling |
| Balkrishna Industries | Deleted | Selling |
Estimates ahead of the event suggested roughly $1.3 billion of passive flows just across the major additions/deletions/weight changes. (Moneycontrol)
But the actual number that matters for today’s market microstructure is much bigger:
₹-equivalent of about $4.1 billion traded on NSE just during CAS.
That was almost 40× the average CAS turnover since its introduction and represented roughly 21% of NSE’s entire cash-market turnover today. (Reuters)
That is enormous.
2. Why Reliance suddenly behaved strangely
This is probably the easiest stock to understand today’s phenomenon through.
Reliance was approximately +0.6% before CAS.
Because MSCI reduced its weight, passive MSCI trackers needed to sell Reliance.
During the auction its indicative price briefly went roughly -1.6%, before finally settling around -0.8%. (Reuters)
So imagine:
3:15 PM
Market says: Reliance +0.6%
Then a huge pool of MSCI-linked sell orders appears.
CAS
Sell imbalance develops → equilibrium price falls.
Final close:
Reliance -0.8%.
Nothing material happened to Reliance’s business during those minutes.
It was essentially portfolio plumbing.
3. Eternal did exactly the opposite
Eternal’s MSCI weight increased.
Before the auction, the stock was down around 3%.
Then the MSCI buying requirement hit CAS.
It recovered virtually the entire fall and closed approximately flat. (Reuters)
So you had this fascinating situation:
Reliance
+0.6% → -0.8%
while
Eternal
-3% → ~0%
during closing-price discovery.
That’s the MSCI rebalance visible almost in real time.
4. Why CAS amplified it
Under the old system, the closing price was essentially based on the VWAP of the last 30 minutes.
Now for F&O-eligible shares, India has a separate Closing Auction Session from 3:15–3:35 PM.
The important mechanics are:
3:15–3:20
Reference-price calculation / transition.
3:20–3:25
Market + limit orders can be entered/modified/cancelled.
3:25–~3:30
Limit orders continue; auction closure occurs randomly within the prescribed window.
~3:30–3:35
Orders get matched and the equilibrium price becomes the closing price.
The CAS price is restricted to ±3% around the reference price, which is based on the 3:00–3:15 VWAP. (NSE India)
And here’s the important concept:
CAS concentrates liquidity.
Previously, an index fund might execute a gigantic rebalance over the final half-hour.
Now a substantial amount of that liquidity wants to meet at one equilibrium price.
That is great theoretically because buyers and sellers meet in one pool.
But if:
MSCI sell requirement = ₹5,000 crore
and natural buyers available at the current price are only:
₹2,000 crore
the auction has to move the equilibrium price lower until enough buyers appear.
That is exactly how you get seemingly absurd indicative moves.
5. What happened to Nifty itself?
This is the subtle part.
Today’s Nifty decline was real, but CAS wasn’t primarily responsible for it.
Approximate picture:
Friday close: 24,175.65
Before CAS today: roughly ~24,050
Final: 24,080.40
So the market had already fallen substantially before MSCI closing flows arrived.
CAS actually improved Nifty by roughly 25–35 points versus its pre-auction level.
The larger day-long fall was mostly macro:
Brent > $90
→ India inflation/current-account concerns
Fed rate-hike expectations ↑
→ global yields ↑
→ emerging-market risk appetite ↓
US–Iran tensions
→ crude risk ↑
HDFC Bank -1.6%
→ major Nifty weight dragging index
IT selling
→ another index drag. (Reuters)
So I would describe today’s tape as:
Weak market + extremely unusual closing-price discovery.
Not:
MSCI caused Nifty to crash.
The most interesting thing today
The CAS actually passed its first enormous institutional stress test reasonably well at the index level.
Ahead of today there were fears it could become chaotic because estimated rebalance turnover was many multiples of normal CAS liquidity.
Instead, approximately $4.1 billion went through the auction, while Nifty’s overall closing distortion was relatively contained. (Reuters)
The problems were much more visible at the stock level.
And that makes sense.
Suppose:
Reliance normally trades gigantic volumes.
A $500m sell order is large, but there is substantial liquidity to absorb it.
But put a $200–500m index order into a less liquid stock and the ratio of:
index flow / available liquidity
becomes enormous.
That’s why the real CAS risk isn’t necessarily Nifty.
It’s individual stocks experiencing temporary price dislocations.
And there is another important wrinkle
CAS currently applies only to F&O stocks.
So even though Groww and Lenskart entered MSCI, they weren’t subjected to the same CAS mechanism, which Reuters noted resulted in comparatively muted auction-related moves in those names. (Reuters)
That creates a slightly strange hybrid market:
MSCI constituent + F&O stock → CAS execution
while
MSCI constituent + non-F&O → normal continuous-market closing mechanism
This will be worth watching as SEBI expands or modifies CAS.
The takeaway for a trader
Between roughly 3:15 and 3:35 on major rebalance / expiry days, I’d mentally separate:
Price movement because investors changed their view
from
price movement because an index fund is forced to transact.
Today, Reliance didn’t suddenly become fundamentally worse at 3:20 PM and Eternal didn’t suddenly become fundamentally better.
The closing prices were partly the result of billions of dollars of mechanical MSCI tracking flows searching for an equilibrium price inside a relatively young auction system.
And remarkably, today’s $4.1bn CAS turnover versus ~40× its previous average may be the most important number from the session—not Nifty’s -0.39%. (Reuters)
I can also track how CAS behaves on the next major expiry/rebalance and flag any similarly abnormal closing moves.