Yet more on the generation game

by John Q on June 5, 2018

Following my critique of generational cliches in the New York Times a while back, I was invited to talk to public radio program Innovation Hub. Here’s the link. If you couldn’t get past the NYT paywall, this gives a pretty good idea of my argument.



Gareth Wilson 06.06.18 at 6:53 am

The Irish abortion referendums seem like a good place to look for real generational effects. A significant change over 25 years, and we already know that over-65s voted No in 2018.


John Quiggin 06.06.18 at 7:46 am

Cultural shifts taking place over very long periods tend to produce this pattern, with the old being least affected for obvious reasons. But there’s no evidence of sharp generational boundaries, and particularly no evidence for boundaries corresponding to the standard US categories. Here’s polling data on Ireland.

As you say, over 65’s (a mixture of Boomers and Silents in the US categorization) are noticably less likely to vote Yes. But even this is just a tendency. Plenty of over 65-s voted Yes, and plenty of millennials voted No. Church attendance (and probably party affiliation, though Irish politics is a mystery to me) would be a much better predictor, I’m sure.


Gareth Wilson 06.06.18 at 9:21 am

” Plenty of over 65-s voted Yes, and plenty of millennials voted No.”
Sure, but we never expect a perfect correlation. “White women voted for Trump” is a meaningful statement about the US election even if all white women didn’t vote for him. It just means white women is a significant category in the electorate. Maybe people born before some year in the 50s is a meaningful category for Irish politics, even if it doesn’t line up with Strauss and Howe’s generations.


engels 06.06.18 at 11:34 am

there’s no evidence of sharp generational boundaries

This tends to be true for a lot of politically salient demographic categories: geography, education, class, ‘race’…


derrida derider 06.07.18 at 12:39 am

I think what your critics here are trying to say, John, is that saying “popular generational cliches are useless” is not the same as saying “all generational analysis is useless”. My schtick, as you will know from past comments, is that the biggest problem with applying generational cliches to public policy is that they regularly confuse cohort and age effects. The old have always been more materialistic and conservative, for instance – its not a characteristic of being born in a particular period (and maybe not even a particular society).

It is also fair to say, too, that applying US generational cliches to other societies is, like similarly applying US racial cliches, especially useless.


John Quiggin 06.07.18 at 4:07 am

@3 Maybe, but there’s no particular reason to think so

@4 I agree entirely. Most of the time apparently sharp categorizations aren’t as sharp as they might appear. All the more reason to avoid imposing spurious categories on variables that are naturally continuous.

@5 Agreed again. I’ve been banging on about the distinction between age and cohort effects for 20 years or so. As I said above, there are obvious reasons why the old will generally be more conservative (in the sense of “old-fashioned”). I’m not sure it’s true in general that the old are more materialistic. But certainly the traits attributed to Millennials and Boomers are just those that have always been attributed to the young and old respectively.


Cian 06.07.18 at 3:22 pm

One reason that the old tend to be more conservative is survivor bias. Generally the wealthier live longer.


Peter T 06.10.18 at 4:47 am

Comments on this entry are closed.