How Late Does This Market Get Round to Pricing a US$1.94 Billion Raise?
The three US indices closed Friday almost where they started. The $Dow Jones(.DJI)$ fell 0.18 per cent to 51,682.64, down 95.40 points on the day; the $S&P 500(.SPX)$ closed 0.17 per cent higher at 7,650.50, up 12.74 points; and the $NASDAQ(.IXIC)$ Composite closed 0.39 per cent higher at 26,522.54. It was a triple-witching day, with stock options, index futures and index options all expiring together and an index rebalancing on top, so volume was always going to be heavy. The calm at index level says very little; what was actually happening was money moving between sectors. The split between the indices is partly a matter of composition, too: not one of the memory names that rose hardest sits among the Dow's thirty constituents.
$SanDisk Corp.(SNDK)$ closed 10.99 per cent higher at US$1,791.82 on heavy volume, having touched US$1,797 during the session, $Micron Technology(MU)$ closed 3.92 per cent higher at US$1,015.80 and SKHY 2.46 per cent higher at US$187.50 — memory is where the money went. That was a second straight day: SanDisk had closed 6.21 per cent higher on Thursday and Micron 5.50 per cent higher the same day, pushed by the Intel chief executive's line that memory prices have risen more than 500 per cent. S&P Dow Jones Indices said on 4 September that SanDisk would join the S&P 100 alongside Dell, Palo Alto Networks and Arista, effective from 21 September, with Nike leaving the index the same day. Friday was the last trading day before that took effect, and the funds tracking the index had to move their positions at the close — part of the gap between SanDisk and the other two memory names is that bid, and it looks at neither the price rises nor 2028.
The reason offered for the rise came from the supply side. Sumit Sadana, senior adviser to Micron's chief executive and formerly its executive vice-president and chief commercial officer, said the current memory shortage covers several markets, and that even with around 20 expansion projects under way worldwide, meaningful new supply will only start to ramp in 2028; on when supply and demand might balance again, the company sees no clear date. On the demand side it is the NAND and enterprise SSDs that AI data centres want, with reports describing supply as short and prices still high.
Part of the day's move, though, did not come from supply and demand. Reports said that SanDisk, Micron, Intel and Marvell saw concentrated buying of short-dated call options that day, with premiums adding up to somewhere between US$93 million and US$96 million, and Friday was the day those options expired. That money is buying a direction over days, which is not the same thing as capacity that starts ramping in 2028.
$Intel(INTC)$, which said it, stayed where it was, closing 0.18 per cent lower at US$108.60. It had closed 7.67 per cent higher on Thursday, and what lifted it was the possibility that SK hynix makes memory chips on its American capacity, not the price rise itself — Intel buys memory, so a higher price is a cost to it. The same sentence is revenue to the side that sells and a bill to the side that buys, and across these two days it is the revenue half that has been priced.
The chip names followed, with $Broadcom(AVGO)$, $Advanced Micro Devices(AMD)$, $Marvell Technology(MRVL)$ and $NVIDIA(NVDA)$ closing between 1.34 per cent and 2.97 per cent higher. None of them had news of their own that day, and the moves were plainly smaller than in the three memory names: what was being bought was the spillover along the same chain rather than the fundamentals of each company.
$NEBIUS(NBIS)$ closed 2.55 per cent higher at US$223.54 and CoreWeave 1.85 per cent higher at US$81.36, having gone live with its Vera Rubin racks that day, while Oracle, which also sells compute, closed 1.98 per cent lower at US$147.61. The two that rose are the ones that had just put new machines in: the money went to capacity already on the floor rather than to the order book.
$Rocket Lab USA, Inc.(RKLB)$ closed 4.79 per cent lower at US$64.57. Its US$1.94 billion equity raise was completed on Thursday, and the shares closed 6.47 per cent higher on the day it landed; the bill for the dilution was only settled on Friday.
The large-cap technology names did not join in, with Meta closing 2.43 per cent lower at US$665.75, Tesla 0.53 per cent lower and Apple 0.26 per cent lower. So Friday's small index moves were the product of a cancellation: what memory and chips added, this group took away. The money did not leave the market, it moved from where the run has already happened to where the price rises have not reached the accounts yet.
The above is personal analysis, not investment advice.
💬 【Talking Point】
$Rocket Lab USA, Inc.(RKLB)$ completed a US$1.94 billion equity raise on Thursday and the shares closed 6.47 per cent higher on the day it landed. On Friday they closed 4.79 per cent lower at US$64.57, with nothing new from the company in between. So how late does this market get round to pricing dilution?
💰 【Bounty】
Drop your view in the comments and there are Tiger Coins in it for you! 🎁
🔔 Better shared than saved — tag a friend and split the coins!
Disclaimer: Investing carries risk. This is not financial advice. The above content should not be regarded as an offer, recommendation, or solicitation on acquiring or disposing of any financial products, any associated discussions, comments, or posts by author or other users should not be considered as such either. It is solely for general information purpose only, which does not consider your own investment objectives, financial situations or needs. TTM assumes no responsibility or warranty for the accuracy and completeness of the information, investors should do their own research and may seek professional advice before investing.
In RKLB’s case:
Thursday: stock rose 6.47% despite the $1.94B share offering.
Friday: stock fell 4.79%, suggesting investors started focusing more on the dilution.
But dilution does not automatically mean the company is worse. Rocket Lab also receives a large amount of new cash to fund growth.
The key question is whether the company can use that money to create more revenue and cash flow than the value given up by existing shareholders.
Simple rule:
New shares → ownership per old share becomes smaller.
New cash → company gets more money to grow.
So I would not judge RKLB only from one day's price movement.
Bottom line: The market can take several days or weeks to fully digest dilution. Watch share count + cash balance + revenue growth + cash flow.
股权融资完成当天，RKLB 还能涨 6.47%，说明那一刻市场交易的可能还是：
融资落地、不确定性消失、公司拿到现金、增长故事继续。
但到了第二天，情绪退下来之后，投资者才重新去算：
新增了多少股份、每股价值被摊薄多少、这 19.4 亿美元未来能不能赚回高于资本成本的回报。
所以我觉得稀释不是“延迟定价”，而是分两层：
第一层先交易融资完成；
第二层再交易融资质量。
如果这笔钱后面能换来更高的发射频率、Neutron 商业化进度、订单和现金流，那短期稀释可能只是代价。
但如果只是不断融资、不断烧钱，而收入和自由现金流跟不上，那市场最终一定会把这笔账重新算回来。
所以我最关注的不是：
“什么时候开始跌？”
而是：
“每增发 1 美元股权，未来能不能创造超过 1 美元的长期价值？”
一句话：
融资当天市场看的是现金进账，第二天开始看的才是股东回报。