Daily newsletter · Mon 14 Sep 2026
| SharespectiveWall Street, before morning tea | Daily newsletterMon 14 Sep 2026 |
Calm On Top Plus: eight foreign ministers meet in Muscat tonight to sign a deal on the Strait of Hormuz. |
Good morning, and welcome to a new week. Last week finished with the S&P 500 down 0.8% and the NASDAQ Composite off 0.7%, which reads like a market that sat still. It wasn't. Four straight down days, a hard bounce on Friday, and underneath those numbers the average American stock had a rough week — because over five sessions traders went from treating a Federal Reserve rate rise as a coin toss to treating it as settled.
That gets decided on Wednesday afternoon in Washington, which is Thursday breakfast here. Before then, tonight, foreign ministers gather in Muscat to sign an agreement on a shipping route through the Strait of Hormuz — the thing setting the oil price, which has been setting the inflation numbers, which have been setting the Fed. A vessel was struck in the strait on Saturday, so nobody expects a quiet signing.
Market snapshot
| S&P 5007,656.98▼ 0.80% | | NASDAQ26,333.04▼ 0.66% | | NZX 5013,580.33▼ 2.80% |
| | ||||
| ASX 2008,741.20▼ 2.10% | | USD/NZD1.7201▲ 1.11% | | USD/AUD1.3945▲ 0.71% |
Levels are Friday 11 September's closes; the change column is the move over the trading week from the prior Friday's close. The NZX 50 and ASX 200 are local closes, and the currencies are New York closes.
| MARKETS |
The index lost less than 1% and the average stock had a much worse week than that
The S&P 500 ended Friday at 7,656.98, down 0.8% over the week. Take the size weighting out and the picture changes: the equal-weight version of the same index fell 1.9%, and the Russell 2000 small-cap index dropped 2.4% to 2,903.94. The Magnificent Seven were roughly flat, enough to hold the headline index up while most of what sits beneath it slid. Across the week futures traders moved from pricing a September rate rise at about 59% to about 87%, and the 10-year Treasury yield added roughly 19 basis points to finish just shy of 5%. Energy led the sectors, up about 1.7%; healthcare lagged at around -3.6%.
Why smaller companies feel a rate rise first. Goldman Sachs reckons about 32% of Russell 2000 companies carry floating-rate debt, against roughly 6% of the S&P 500. Floating-rate debt is a loan whose interest resets with the market, so a central bank move turns up in the interest bill within months rather than whenever the next bond is refinanced. Bank of America's estimate is that every 25 basis points costs the Russell 2000 about 2% of operating earnings — a cash cost, not an argument about valuation.
What to watch the gap between the index and the average stock. It has widened all year, and Wednesday's projections decide whether it keeps going. |
| HEALTHCARE |
Amgen had its worst day since 2000 because somebody else's drug failed
On Tuesday, Amgen fell 10.08%, the worst performer in the Dow and its steepest one-day drop since October 2000. Nothing had gone wrong at Amgen. Novartis had announced that pelacarsen — its drug for lowering a blood particle called lipoprotein(a), or Lp(a) — missed the main goal of its late-stage trial. Patients taking it had no meaningfully fewer heart attacks, strokes or cardiovascular deaths. The drug lowered Lp(a) as designed. It just didn't appear to help.
Why that became Amgen's problem too. Lp(a) is inherited, barely shifts with diet or statins, and high levels have been tied to heart disease for decades, which made "lower it and see what happens" one of the better ideas going in cardiology. Amgen has its own candidate, olpasiran, which switches off the gene that produces Lp(a) and cut levels by more than 95% in mid-stage testing. If lowering Lp(a) doesn't prevent heart attacks, lowering it a great deal more doesn't obviously prevent them either.
There is a real counter-argument: Novartis's drug cut Lp(a) by far less, and the levels it reached may not have been low enough to matter, which makes Amgen's 2028 outcomes trial a different test rather than a repeat. Amgen closed Friday at US$377.35, about 15% below its 52-week high.
The open question whether Lp(a) causes heart attacks or merely marks out the people likely to have them. That gets answered in 2028, and it got priced on Tuesday morning. |
| SOFTWARE |
The week's best and worst stocks were the same AI trade pointing in opposite directions
The five best performers in the S&P 500 last week were Lumentum, up about 9%, Hewlett Packard Enterprise at 7.2%, TKO Group, Edison International, and Ciena at 4.2%. Three of those five sell physical equipment into datacentres — optical modules, servers, networking gear. At the other end sat two software companies: FactSet, off about 14%, and PTC, off about 12.6%.
What the market thinks AI does to a software company. FactSet sells data and analytics terminals to investment professionals on subscription, and the fear is easy to state: if a general-purpose model can read the filings and build the spreadsheet, a specialist terminal is less special. The company's answer is that it has six AI products in front of paying customers and expects them to add 30 to 50 basis points of growth this year, and it has narrowed full-year subscription growth guidance to 4.4% to 5.8%. The shares have more than halved over twelve months anyway.
FactSet is not an isolated case. Software has spent 2026 de-rating as a group, frequently while reporting respectable results, and the money has gone to whoever supplies the picks and shovels instead. Lumentum makes the optical parts that shift data around a datacentre; Ciena makes the networking kit underneath. Neither has to argue about whether a model might replace it.
The other side nobody has yet produced a large software company visibly losing customers to a chatbot. Until someone does, this is a story about what investors fear rather than what has happened — and fear reprices faster than revenue does. |
Newsworthy numbers
| | |
| 5.36% Australia's 10-year government bond yield at the end of the week, its highest since May 2011. Markets put the odds of the Reserve Bank lifting the cash rate to 4.60% on 29 September at roughly 80%. | | €3.82bn Tata Motors' all-cash offer for Italy's Iveco Group, open until 26 October. An Indian carmaker buying one of Europe's great truck builders is not how anyone drew this up. |
| | ||
| | |
| A$630,000 what the super industry's own benchmark says a single Australian homeowner now needs at 67 to retire comfortably, up from A$595,000 and the first rise in three years. | | 66.5m years the age of the first trackway ever found from a fully grown Tyrannosaurus rex, four prints in the North Dakota badlands, each about a metre long. The spacing suggests it was walking at about 4mph, in no particular hurry. |
Worth a read
| |
| 01 | Hope for spark to ignite the embers in June quarter GDP figures — Thursday's New Zealand GDP print, previewed. The banks have quietly upgraded: ASB at 0.3%, Westpac and BNZ at 0.2%, Kiwibank and ANZ at 0.1%, against a Reserve Bank pick of exactly nothing. Construction rose 4.8% in the quarter, which nobody saw coming. |
| |
| 02 | September changes to hit millions: Centrelink, Age Pension, rent allowance and travel rules — Rates move on 20 September: the single Age Pension up $36.80 a fortnight to $1,237.70, JobSeeker up $16.20 to $833.70. The quietly useful change is the travel rule, which lets retirees spend 12 weeks overseas instead of six without losing the Pension Supplement. |
| |
| 03 | From toilets, textiles and MSG to AI semiconductor chips — Toto makes the ceramic chucks that hold silicon wafers still, and Ajinomoto, the MSG company, makes an insulation film found in about 95% of high-performance chips. None of them planned any of this. |
| |
| 04 | Big supernova dataset challenges dark energy theory — A University of Queensland-led team stitched three decades of observations into a catalogue of 2,884 Type Ia supernovae, and found more evidence that dark energy changes over time rather than holding constant. The standard model is having a difficult year. |
| |
| 05 | Can Andre Agassi enjoy pickleball more than he hated tennis? — Lizzy Goodman in The New York Times Magazine on a man who was the best in the world at a sport he did not enjoy, now working out whether he can enjoy a smaller one. He is not sure, and says so. |
Wrapping up
Last week was a slow repricing — four sessions down as the rate-rise odds climbed, one up when oil finally fell. The Federal Reserve decides on Wednesday 16 September at 2pm in Washington, which is 6am Thursday in Auckland and 4am in Sydney, with about 85% odds priced on a quarter-point rise to 3.75–4.00% and fresh projections and a dot plot alongside it. The Bank of England follows on Thursday with only a 25% chance of a move priced, and the Bank of Japan on Friday, where a rise to 1.25% is roughly 90% priced.
The data runs through the decision: August retail sales at 8.30am New York time on Wednesday, housing starts and the Philadelphia Fed index on Thursday, industrial production on Friday. Earnings are thin. Trip.com reports after Tuesday's close, where the line to watch is accommodation and packaged-tour bookings — the cleanest read available on whether travel demand holds up with fuel this dear — and Lennar follows after Wednesday's. Locally, New Zealand's current account lands on Wednesday, June-quarter GDP on Thursday and August trade figures on Friday; Australia gets the Westpac-Melbourne Institute leading index on Tuesday, and the RBA does not meet until 29 September.
The kiwi was the weakest of the major currencies last week, down 1.1% to US$0.5814 and taking USD/NZD to 1.7201 in a third straight weekly decline, while the Australian dollar lost 0.7% to US$0.7171 and left USD/AUD at 1.3945. Both were run over by a US dollar firming on the Fed trade, which for now counts for more than the RBNZ and the RBA being expected to raise rates of their own.
Have a good week — back tomorrow with tonight's session.
General information and commentary only, and not financial advice. It does not take account of your objectives, financial situation or needs. It reflects the author's own reading of public information at the time of writing and may be wrong. Do your own research and talk to a licensed financial adviser before making any investment decision.