AI stocks led the sharpest one-day fall since April 2025. New Zealand has already run the experiment on what panic-switching costs, and the answer is on record.
I've been quiet for a few weeks while I finished the New Zealand Certificate in Financial Services (Level 5), Investment strand. It's done and passed. Now I'm back to it, and the market picked a good week to give me something to write about.
On 5 November the Nasdaq fell 5.5% in a day and the S&P 500 fell 4.4%, both their worst days since April 2025. Nvidia dropped more than 9%. The seven biggest US tech companies lost over US$1 trillion in value between them. The trigger was a worry that the AI data-centre build-out would slow after the US midterm results, and the selling landed hardest on chipmakers and data-centre names. By the following Monday the Nasdaq was more than 10% below the record it set in October, which is the technical definition of a correction.
From here it looked a little softer. The NZX 50 dipped but held up far better, and the NZ dollar slipped from about 56.2 US cents to 55.5 over the same couple of days. For anyone holding US shares unhedged, that softened the fall slightly in New Zealand dollars. Currency cuts both ways, as Issue 22 showed, but this time it was on our side.
My own disruptive innovation position, with the AI layers on top of it, is about as close to the centre of that sell-off as you can get. Chips were where the selling started.
I did nothing. I didn't sell, I didn't try to buy the dip, and my regular contributions went in the same as any other week. That wasn't willpower. The decision about what happens on a bad day was made months ago, when nothing was falling, and all I had to do was leave it alone.
One email a week. Straight writing on investing from New Zealand.
New Zealand has already run this experiment, and the results are on record.
When markets crashed in early 2020, KiwiSaver members switched funds in a hurry. Research commissioned by the Financial Markets Authority found $1.2 billion moved into lower-risk funds during the COVID market fall, and only $121 million moved back. Around 12,700 members aged 26 to 35, people with three or four decades before they could touch the money, switched from growth funds to conservative ones between February and April 2020. A year later, more than 12,000 of them were still there.
The S&P 500 fell about 34% between 19 February and 23 March 2020. On 18 August 2020 it closed at a new record high. Under six months. The people who switched to conservative sold after the fall and missed the recovery. They turned a temporary loss into a permanent one, and that was the only part of the whole episode they controlled.
It isn't true for every dollar. Short-term swings matter for money you need in the short term. Saving for a house deposit you plan to use next year, or drawing down at 65, is a different situation, and Issue 19 covered why the order of returns matters so much once you start withdrawing.
The difference is that the decision gets made before the fall, not during it. Money you'll need within a few years shouldn't be in growth assets in the first place. Money you won't need for decades should be able to sit through a day like 5 November without anyone touching it. If a bad week forces a change to your plan, the plan was set up for a different timeframe than the one you're on.
Keep the regular contributions going. Every month you keep investing through a fall buys more units at lower prices. That's the same dollar-cost averaging Issue 13 made the case for, working in your favour when it feels least comfortable.
Check the size, not the price. Issue 20 argued a satellite position should be a few per cent of the portfolio, not a quarter of it. If you sized it that way, a 9% fall in Nvidia is a small dent in the whole portfolio. If a single day in one corner of the market ruined your week, that's a sizing problem to fix calmly, not a reason to sell in a hurry.
Check less often. A KiwiSaver balance you won't touch for thirty years doesn't need checking on a day the Nasdaq falls 5.5%.
And if you still want to change something, write down why and wait a week. If the reason still holds once the headlines have moved on, it might be a real reason. Most of the time it won't be.
Nobody knows whether 5 November was the start of something bigger or a bad day that gets forgotten by Christmas. That's the point. A long-term plan doesn't need to know.
One email a week. Straight writing on investing from New Zealand.
Subscribe for free