In 1973, 843 people died on New Zealand roads. Last year, with far more people driving far more cars, the provisional toll was 272.

Nobody held a press conference, and why would they? The improvement stretched over so many decades that hardly anyone noticed it, and nobody could claim it as their own.

The toll did not fall because of one heroic decision. It fell because safer cars, restricted licences for new drivers, tougher drink-driving enforcement and better road engineering each shaved a little off the danger over half a century. Every step was small, sensible and politically painless. None required anyone to be brave.

There are more such good news stories in the long-term data. With my colleague Bryce Wilkinson, I have just compiled it all for a new book. New Zealand by Numbers traces more than a hundred measures of New Zealand life back to 1970 and beyond.

Life expectancy, for one, has risen almost eleven years since 1970. About 36% of adults smoked in 1976, against fewer than 7% daily today. Of every 1,000 babies born in 1970, almost 17 died before their first birthday. Last year, fewer than five did.

But some charts in the book refuse to behave, which means some problems persist or even get worse over the decades. And there are reasons why some indicators keep improving while others do not.

The improving charts have something in common. Small gains pile up on each other, like compound interest.

The stubborn charts exhibit the opposite. In 1970, an hour of work in New Zealand produced almost exactly what an hour of work produced in Britain. Today, the comparison is embarrassing.

Economists compare countries in dollars adjusted for what they buy in each place. Australia now squeezes 77 dollars from every hour of work, Britain 68 and New Zealand not quite 50. An Australian hour produces more than half as much again as a New Zealand one. While small productivity gains have accrued, the gap has kept on widening.

Some problems have also compounded. The Infrastructure Commission found that before the Town and Country Planning Act 1977, house prices rose about 0.5% for every 1% of population growth. After the Act handed objectors a development veto, prices rose 2% for every 1%.

The Resource Management Act in 1991 made a bad situation worse. Nobody intentionally legislated a housing crisis. It simply built up one upheld objection and one district plan at a time. The country and its geography have not changed. Only the rules have.

Productivity and housing defy solution by small steps. Fixing them means upsetting the way things are done, and upsetting people is something New Zealand politicians rarely want to do.

It was not always so. The reforms of the 1980s and early 1990s were among the boldest in the developed world, and the productivity chart show how well they worked. New Zealand partially caught up through the 1990s as the reforms took hold. Then the catching-up stopped, at roughly the moment reform did.

The country looked back at its one great burst of courage and drew a lesson. It was not that the reforms had failed but that they had hurt at the time and cut short a few political careers along the way. The lesson was “never again.”

Ministers since have preferred managing decline rather than confronting it. That is why Erica Stanford stands out. New Zealand’s PISA mathematics score has fallen 58 points since 2000, roughly three years of learning. Among all the statistics we went through, this was one of the most depressing ones.

Stanford’s response has been the opposite of managing decline. She has overseen the writing of a new curriculum, put structured literacy and maths into every primary classroom, and set about replacing NCEA. From 2028, students will move to new qualifications with marks out of 100 and grades that parents can understand.

Agree with every detail or not, this is reform in the 1984 mould: big, fast and deliberate, designed to put a visible kink in a chart rather than a gentler slope.

We will not see the reforms’ full effects for years. The next PISA round will mostly test children taught under the old settings. And unfortunately, there is another problem. New Zealand has a habit of giving up on reforms, or reversing them, before they have had a chance to work. The question is not whether boldness pays, since the 1990s answered that, but whether voters will let it run long enough to compound.

The road toll, meanwhile, will keep falling. Cars will keep getting safer, perhaps one day driving themselves, and the decline will continue without anyone fighting for it.

Nobody will hold a press conference then either. That is a comfortable kind of progress, because it asks nothing of anyone. The uncomfortable kind must be chosen, again and again, by ministers willing to be shouted at.

Our book records both the progress that needed no courage, and the problems still waiting for it.