Late-day buying was observed in Chinese stocks this week, with notable activity in the final 20 minutes of trading. The focus was on ETFs preferred by China’s state-run agencies, known as the National Team.
Increased trading in Huatai-Pinebridge CSI 300 ETF and China AMC SSE 50 ETF contributed to the Shanghai Stock Exchange Composite Index advancing for an eighth consecutive day. This marks the longest winning streak since October.
Upcoming Market Opening
US President Trump’s positive remarks about a potential China deal could influence the upcoming market opening. This development reflects Beijing’s efforts to support local markets during challenging times.
The late-session inflows highlight a pattern we’ve often seen when domestic institutions step in to stabilise sentiment. With shares rising steadily across consecutive sessions, there’s a clear attempt to restore confidence precisely when liquidity tends to thin — just before the market closes. Such timing is hardly accidental. When large purchases cluster in the closing minutes, they can nudge broader benchmarks upward without drawing much attention earlier in the day, and this has implications for short-dated contracts tied to large-cap indices.
The concentration in particular ETFs – those tracking CSI 300 and SSE 50 names – isn’t just a coincidence either. These funds are heavily weighted in state-owned and strategically positioned companies, which tend to act as bellwethers. By anchoring demand here, the moves ripple outward through futures linked to these indices, narrowing the bid-offer spread and tightening intraday ranges.
Li’s administration appears committed to policy steadiness, even as foreign fund flows display hesitation. When Trump’s comments lifted sentiment, it amplified existing movements domestically rather than sparking them. The mood, in other words, was already turning before the external catalyst came into view. That reflects a strategy that relies less on reactive changes and more on quiet reinforcement from within.
For us, what’s material is the traction seen in implied volatilities. Levels on shorter-term contracts have faded alongside these late-day climbs, suggesting that rapid price jolts are being priced out by market makers, at least provisionally. Option skew has flattened, which normally tempers demand for protection as downside expectations pull back.
Changes in Volume
It’s worth pointing out how these price dynamics affect rolling strategies on near-term futures. As directional bias grows clearer, positioning tends to move more purposefully – even though the broader conviction may still be lacking at the sector level. Long gamma strategies may come under pressure as realised swings decrease, while delta-neutral trades would benefit from the narrower bands.
We’re also seeing subtle changes in volume build-ups during the midday session, a period that used to be largely void of interest. Now, there’s more two-way interest, which could suggest that yesterday’s patterns are informing today’s open, not just the close.
Across the week ahead, correlation among sector indices has ticked lower. That often softens the need for broad index hedges and gives more room for dispersion trading to function. We may also adjust our exposure around calendar spreads depending on how Friday closes unfold, since weekly flows have been more lopsided than usual.
Zhang’s policy cues continue to point to steady absorption rather than intervention through sharp changes. As such, price readings in the next few days should be weighed against both direction and pace – a tradeable mix when bracketed efficiently.
Keep an eye on the shifts in ETF premiums relative to their net asset values. These have diverged, particularly at the top end of the trading day. This pattern gives us hints as to whether retail money is joining or simply mirroring the larger flows. When these premiums return to neutral levels quickly, it often hints at more disciplined order flow rather than speculative bursts.
We’ve adjusted our models this week, nudging hedging ratios modestly lower in response to realised ranges. That decision comes not from any single event, but from data that’s become harder to ignore.
