Here are two numbers that don't belong in the same country. In 2000, New Zealand recorded 7,650 serious non-fatal injuries. In 2008, it recorded 7,815. Nine years, and the count moved by 165. Then look at the most recent available figures, from 2024: 14,283 (Source: Stats NZ / ACC, fatal-serious-injuries).

A near-flat decade, followed by something close to a doubling. That is an 87 percent rise on the 2000 figure, and it did not happen in a single bad year. It happened relentlessly.

The turn came in 2009, when the count first broke 8,000 at 8,238. Every year since has been higher than the one before it. That is 16 consecutive annual increases, with no down year anywhere in the run. Through recessions, through a housing boom, through a pandemic, through a construction slowdown, the line only ever went one way.

Now the part that actually matters for where we're heading. The steepest stretch was 2015 to 2019, when the count went from 9,861 to 12,945. That is an extra 771 serious injuries a year, every year, for four years. The five years since have been very different: 2019 to 2024 added 1,338 in total, an average of about 268 a year. The 2019 to 2020 gap was just 63.

So the record high and the slowdown are both true at once. The count keeps setting new peaks, but the engine driving it up has lost most of its power. The most likely reading is that whatever changed in the late 2000s, whether that was exposure, recording, or genuine risk, has largely worked its way through the system, and we are now sitting on a high plateau rather than a climb.

The honest alternative reading: these are counts, not rates. New Zealand's population has grown substantially since 2000, and it has aged, and older people are more likely to suffer a fall that meets the serious injury threshold. A count that rises while the population rises does not automatically mean the average person is more likely to get hurt. This table doesn't include a rate, so it cannot settle that question. Nor can it rule out changes in how injuries are classified or captured over 24 years, and the most recent year in any injury series is the one most likely to be revised. Any of that could explain a real slice of the increase.

What it cannot explain away is the size of the number. 14,283 serious non-fatal injuries in a single year works out to roughly 275 people a week. Not bruises. Injuries severe enough to be classified as serious: the ones that end careers, empty savings accounts, and reshape a household's next decade.

My view, and it is a view rather than a finding: flattening out just below a record is not a safety success story, and it shouldn't be sold as one. A curve that stops climbing is only good news if the next thing it does is fall. Sixteen years into this run, that hasn't happened once. The 2000 to 2008 data proves a stable number is achievable. Nothing in the last 16 years proves a falling one is.