Then, the relative size of a bottom or top absolute threshold is highly meaningful. Even if it's a fraction of a percent, populations are huge and suicide rates are not rounding errors at all -- they're actually quite statistically significant.
And as macabre as it is, suicides are objective facts mostly unaffected by methodology, and unaffected by translation issues, cultural differences, etc.
This is why suicide rates are actually a powerful mental health statistic, just like height is a powerful physical health statistic, at the population level. There's obviously still a lot both of these metrics don't say, but the fact that they are highly objective makes them extremely valuable.
"The large variations in the systems and processes to define mortality causes imply there may be very different numbers of deaths that are registered with a specific cause. This creates a problem for cross-country comparisons of mortality by cause in general, and even more so for deaths of despair, and suicides in particular.
The person responsible for writing the cause of death on the death certificate may be different across countries. In some countries, the police are responsible, while in others a medical doctor, coroner, or judicial investigator takes on this role. Differences in doctors’ training, access to medical records, and autopsy requirements contribute to these discrepancies. The legal or judicial systems that decide causes of death also vary. For instance, in some countries suicide is illegal and is not listed as a classifiable cause of death, leading to underreporting or misclassification of suicides as accidents, violence, or deaths of “undetermined intent.”[25]
Data on suicides, even when reported, can be inaccurate due to social factors as well. In some countries, suicide might be taboo and highly stigmatised, so the families and friends of the person who committed suicide might decide to misreport or not disclose the mortality cause, causing underreporting of its incidence. In other societies, such as Northern Europe, there is less stigma attached to suicides, and alcohol and drug use."
https://www.worldhappiness.report/ed/2025/supporting-others-...
I wouldn't be surprised if cultural differences are actually the largest factor that explains a country's suicide rate. Not easy to prove, of course, but I would be very careful drawing any conclusions from differences in suicide rates between countries with vastly different cultures.
I think you can also expect large differences in how countries report their suicide rates.
That's a pretty strong assumption, seems more likely that there's variation at the extremes than not. For example, if a small percentage of the population deals badly with extended nighttime in long winters, then it'll affect Finland's most-unhappy stats (and suicide rates) without meaning much for the average happiness.
Prove it
[1]: https://www.mhlw.go.jp/stf/seisakunitsuite/bunya/hukushi_kai...
The fact that people want to change their culture is possibly an early indication of a shift, which could take decades or centuries to actually occur. And such a cultural shift can also lose momentum and be still-born.
---
I find counting suicides innovative. But if you do it in a global context without looking at the cultures as confounding factor: It's wrong.
There are many other confounding factors, such as a forgiving national (personal) bankruptcy regime. The USA has a pretty forgiving regime compared to other countries. But that doesn't mean you can say it correlates with how happy people are. Because - like suicides - the number of people that go bankrupt might not significantly correlate to the average happiness rate. Because a (small) minority of people go bankrupt / commit suicide.
It's in fact perfectly reasonable and possible to suppose that a country with higher average suicides and harsher penalties for bankruptcy still ends up higher on the happiness index. Because perhaps health and social-contact / family factors impact the rating more, on average.
If people are happy, you have less suicides. I don't need a study for that.