That’s the baby boom moving up the chart.
Yes, exactly my point. The boomer generation itself made the population pyramid look different at every stage of its life, which is why the 1980 chart looks so different from the 2023 chart. When you introduce a cohort that has its own slope from birth statistics, the shape of the drop off at 60 is confounded by the preexisting shape of the slope before they entered old age.
So the appropriate method of isolating the variable that shows what you call a “die off” would be to just pull up the actuarial tables that show what percentage of 60, 61, 62 year olds, etc., die that year. Not to compare how many of those there are as a percent of overall population.
You keep calling it a “die off” because you’re being visually tricked by the misleading population pyramid. Use the actuarial tables instead.
Among 65 year old men, the probability of surviving to 75 is 76%. The probability of surviving to 85 is 39%. The probability of surviving to 95 is 5.9%.
For women, the odds are 84%, 52%, and 12% of getting to 75/85/95, respectively.
Yes, these are higher death rates than at younger ages. But nowhere near what the shape of the population pyramid suggests, where the 85 age cohort is about 1/4 as large as the 65, which misleadingly suggests a probability of 25% of living 20 more years, when the real number is closer to 45%.