*The Great Escape*

The author is Angus Deaton and the subtitle is Health, Wealth, and the Origins of Inequality.  It is a very good book, as you might expect.  Here are two bits I found especially interesting:

In Sweden in 1751 — well before the modern mortality decline — it was riskier to be a newborn than to be an 80-year old.

And, somewhat more recently:

…until around 1900, adult life expectancy in Britain was actually higher than life expectancy at birth.  In spite of having lived for 15 years, those teenagers could expect a longer future than when they were born.

The book’s home page is here.


Are either of these particularly surprising? Conditional probability at work. All it says is that infant and child mortality was high enough relative to death from other causes in adulthood that once you reached adulthood you had a higher life expectancy due to the removal of the set of people that died young. Might be a little surprising to talk about the Swedish 80-year-olds but certainly not for adults vs infants. The latter disparity has existed for most of human history.

What surprised me was that we have good enough data for 1700's Sweden to be even able to make a claim like that.

It is scary how good parish records in Sweden are. You can actually take that data reliability into the 1620s even in particularily remote areas. I had a relative, in Sweden, who decided to trace every single ancestor, the only ones who couldn't be traced to the late 1500s were a couple of Sami women who were very likely not from Sweden. Even displaced famine refugees from Finland and a whole line of half Sami reindeer traders were completely recorded with birth, marriage, and death dates. The least reliable data came from young men who died on military service, and even here you could find when their names were struck from the regimental lists.

"it was riskier to be a newborn than to be an 80-year old."

Its such a fine line between clever and stupid

But it is not a fine line between the risk of living 40 more years for the infant, compared to the certainty that the 80 year old wouldn't live to 120. Not to mention the fact that the 80 year old started out with the same 'risk' as all the other infants.

Somehow, it is not a surprise that an interesting example of oblivious observation which misses the point is highlighted here.

I expect that by 'riskier', he means that one-year mortality for newborns was higher than it was for 80 year olds?

The alternate interpretations would be that he is saying that median life expectancy is lower for newborns (implying that a majority of newborns die in infancy/toddlerhood, given the likely life expectancy of an 80 year old was probably at most 5-6 years.), or that mean life expectancy is lower (meaning that so many kids are dying that the very few who live a long life aren't enough to offset all the dying infants). The median interpretation is unlikely and the mean interpretation is ludicrous outside the context of the first born of Egypt in Moses' time.

Yes, but I suspect it's reversed in the 20-year mortality rate.

I wonder if many children working in very dangerous and dirty factories during the first and second industrial revolutions is a significant reason for the short life expectancy for the young. I don't think parents waited until their child was 15 to send them to work in a factory (especially during the first industrial revolution). I recall seeing pictures of 6 and 7 year-olds working in factories.


I wonder if many children working in very dangerous and dirty factories during the first and second industrial revolutions is a significant reason for the short life expectancy for the young

Seeing as the numbers are from 1751 and the industrial revolution started in Northern England c. 1760, I'd say no.

Have you seen these people? Toddlers are idiots. Letting them run machines is probably what made the factories dangerous.

Also, infant mortality throughout most of history was far more dangerous than any factory outside of a concentration camp.

survivorship bias....


Sorry, Tyler, but I don't see what's interesting about those 2 tidbits.

It's well known that, before our era, it's the first 5 years of lives who were the most dangerous. I can still recall an old History teacher explaining to us that 40 years old life expectancy was an average. He used the comparison with a minefield. In the first 5 years, you had to cross a field densely populated by mines and so dying was extremely easy/common but, as you aged, the mines were more spread out, less concentrated so your survival prospects, IF you had survived so far, improved.

Today, if you have kids, you can still realise this truth. Before the kids are 5 yo, it likely seems to you you're going to the doctor every week, every other week at best. And that kids are always having some kind of fever or something. And it does get a lot rarer once that 5 years milestone is being crossed.

What's today is just a bit scary - a kid with 39-40C temperature - was often deadly in earlier times...

It does seem unlikely that anyone reading this book won't already get those facts. But in the general population I don't think you'll find a ton of people that would get this without explanation.

There's an email clogger going around that's a list of all the changes in the world (meaning the U.S. and Western Europe, of course) since 1910 (one of those "shocking true facts!" lists), and one on the list is exactly what you note, the life expectancy of 47.

Population statistics really seem to fit into the "there's lies, damned lies, and statistics" model because they're so easily used and few seem equipped to call bologna on them. I've heard many times recently, for example, the talking point that this generation of young people is the first with a projected life expectancy shorter than the generation before them. This is an absurd statement to make without qualifications concerning which population you are including (and it's sketchy even if you limit it to middle class Americans of this century, since it's all conjecture). But it gets repeated over and over again as if it's common sense.

NB: I am not saying this is a bad book: I had a look at the homepage/Amazon page and, though I don't think I will buy it as it doesn't seem to say too many new things, it does seem interesting.

I would agree that international aid, so far, doesn't seem to have shown much results. So new ideas on how to achieve lift-off are, of course, interesting.

But, again, liberalising trade in agricultural products and stopping western countries from subsidising their own agricultural sector is pretty old hat and has no chance of being accepted politically.

Does Angus Deaton propose other things?

So the upshot is that the last few centuries have seen the largest, most unprecedented, intergenerational transfers of health that the world has ever seen?

Next time someone is complaining about how their infant will never get social security maybe I can remind them of how much worse it could be...

Well, not transfers. It's not as though the old have gotten sicker so that the infants could be healthier.

Unless you're parodying the leftists who say that the rich getting richer inherently makes the poor worse off even if they're getting richer, too. In which case, good show.

For the United States currently, the expected years of additional life for children are: newborns, 77.3; 1-year-olds, 76.8; 5-year-olds, 72.9; 10-year-olds, 67.9.
(from Table A on page 3, www.cdc.gov/nchs/data/nvsr/nvsr53/nvsr53_06.pdf)

The drop of life expectancy of only half a year for 1-year-olds comes about by 0.7% of newborns not reaching their first birthday. For newborn black Americans, 1.4% don't reach their first birthday, and so expected years of additional life goes up a bit from 72.3 years for newborns to 72.4 for 1-year-olds. Even a small fraction of babies and children dying significantly attentuates average life expectancy of the young, since those dying young die so many decades before their cohort average.

Robert Fogel covered a lot of similar ground in "The Escape from Hunger and Premature Death, 1700-2100." Deaton reviewed Fogel's book when it came out and pinpoints the scholarly disagreement between them: Deaton gives more credit to medicine and public health while Fogel insists nutrition (and especially the nutritious diets available to pregnant mothers) is almost completely responsible.

Review here: http://www.nber.org/papers/w11308.pdf

Doesn't surprise me a bit. My wife is due in a month, and belongs to an online moms group with tens of thousands of others who are also due this month. So many heartbreaking stories...

You become acutely aware that your child is constantly a few minutes from death from lack of oxygen. Of course we all are to some extent, they really need to get to work on those respirocytes.

Comments for this post are closed