What has changed in the lives (and pattern of death) of American women and men?

I was so much intrigued by Paul Krugman’s tweet that I thought I would dig a bit deeper. Knowing that life expectancy differs significantly between men and women, I wanted to see how the picture would look like if one looks at the same difference to the OECD average, but by gender. And I must say I was on for a surprise.

Dear friends in the United States: can you tell me what story may be behind this?  Yes, the title is a real question!

Let me expose the situation I found through the data on life expectancy by gender comparing the evolution between 1960 and 2015 in the United States and on average among OECD countries.

In a previous post, we have seen how the evolution of life expectancy is a critical element in the progress of aging in the American context, as it is, but plays differently, in each national context. I used the same data as the ones reported in Paul Krugman’s tweet and the Washington Post article.

 

Life expectancy at birth

I provide a first set of charts through the short slide show below. Click on the arrow to the right and enjoy the build-up of the charts.

Chart 1 provides the full evolution of life expectancy at birth by gender, i.e. how many years a newborn girl or boy in a given year can expect to live. It shows how fast life expectancy at birth has increased and how different the evolution has been in the United States compared with the average across OECD countries (including the United States) – giving exactly the same representation as in the Washington Post article, but for women and men separately.

Chart 2 just takes these data a step further and portrays the difference in years of life expectancy at birth between the United States and the average for OECD countries.

• When life expectancy at birth in the US is higher than that for OECD countries on average, the bars are above the horizontal zero line.
• When life expectancy at birth in the US is lower than that for OECD countries on average, the bars move below the horizontal line.

 

 

Now, let me give my descriptive remarks on what we see in these charts:

First, from Chart 1 : on average among OECD countries, life expectancy at birth has always been higher among women than among men – this is a general pattern across all countries –, the two lines move up largely in parallel. In fact, in the first part of the 55-year period, the gap increased by about 1.5 year, from 5.3 years in 1960 to 6.8 years in 1979. It remained unchanged until the mid 1990s. Then the gap shrunk regularly to come back to 5.3 years in 2015.

Second, life expectancy at birth also increased constantly (or almost) year after year in the United States for both women and men, but the increase has been far from linear. For women, life expectancy at birth has been above the OECD average until 1994, but after 1982 the increase in life expectancy at birth for American women had much slowed down, resulting in their longevity becoming lower than on average in OECD countries after 1994. The gains in life expectancy at birth among American men crawled more closely along the OECD average with a somewhat positive gap created between 1973 and 1993. After 2004, about ten years later than for women, life expectancy at birth for American men moved below OECD average.

Third, with Chart 2, we now consider the difference in life expectancy at birth between American men and women and the respective averages across OECD countries – the vertical difference between each pair of lines in Chart 1. The first striking feature is that the magnitude of the difference between the United States and the average for OECD countries is always much larger for women than for men. This happens both when the comparison is positive for the United States (life expectancy at birth higher in the United States) and when the comparison turns negative, after 1993 for women.

Finally, the negative gap after 1993 for women and after 2002 for men has been ever increasing and does not show any sign of getting stable.

 

Life expectancy at 65

In a second set of charts, built like the first one, I look at life expectancy at age 65, i.e. the number of years a person aged 65 can expect to live considering the conditions of life and death in the given year. This considers only persons who reach the age 65 – it means that it is independent of any variation of death rates for ages below 65. Click on the arrow to the right and enjoy the build-up of the charts.

 

 

Similarly to the first set of charts, Chart 3 provides the full evolution of life expectancy at 65. It shows how fast it has increased and whether the evolution has been different in the United States and on average across other OECD countries.

Chart 4 adopts the same approach as Chart 2, presenting the difference in years of life expectancy at 65 between the United States and the average for OECD countries.

These charts call for additional remarks:

First, as one reads on Chart 3, life expectancy at 65 is much higher for women than for men. Progress of life expectancy at 65 has been quasi linear on average for OECD countries during the whole period covered (1960-2015) for women, gaining about 1.1 year every ten years. Such a linear pattern does not show up for men: life expectancy at 65 remained stable around 13 years for the first 15 years of the period, then started to move up at a faster pace than for women (1.2 year every ten years).

Second, life expectancy at 65 has been significantly higher in the United States than on average across OECD countries for both men and women, during most of the period covered. But, after 1992, further annual progress paused in such a significant way for women that by 2001 it had become lower than the OECD average. When progress then resumed, it did not quite keep pace with the OECD average, creating a slowly growing negative gap. The pattern of evolution is not the same for men: life expectancy at 65 for the United States moves up constantly and close to linearly, always above the average for the OECD countries. The line never (at least until 2015) goes below that of the OECD average.

Finally, Chart 4 emphasizes these different patterns for men and women. The difference between the United States and the OECD average is much larger, positive, for women than for men, until it decreases sharply and turns negative. For men, the difference always remains positive after 1968. While still positive, it gets to its lowest level ever since 1968 in the last years of the period, i.e. coming to 2015.

The question haunts me:

 

What has changed in the lives (and pattern of death) of American women and men?

Not only the overall pattern of evolution of life expectancy at birth in the United States breaks away from the general evolution in other OECD countries (taken together rather than compared individually), but also such patterns appear clearly different by gender. When considering both life expectancy at birth and life expectancy at 65, the nature of the evolution against the OECD average is different in magnitude. The changes in slope of progress is different in timing and magnitude. If it often results in lower longevity in the United States than on average among OECD countries, it is not the case, at least yet, for added years of life after 65 years for men.

What does explain these rather drastic changes in the evolution of life expectancy in the United States – their timing, their noted differences in timing and magnitude by gender? I have come across some possible explanations like changes in the ethnic composition of the population with different life expectancy patterns or the increase in violent deaths. However, I do not see how and why such underpinnings would explain either the timing or the different evolution by gender in life expectancy, highlighted in this post.

Can I count on readers of this post – of course, most likely, among my American friends – to shed light on this question or share any knowledge or interpretation? I’ll be very happy to share back such contributions here.

---

Non solum data – Data sine monito oculo nihil sunt.

---

Through this blog, we invite constructive comments and constructive article contributions – review our blog policy and have your say!