Registration tables vs. Mortality tables

American politics right now is basically a race of the registration tables against the mortality  tables.

Charles Franklin has a great graph on Twitter looking at mid-term voter participation rates by age for 2010, 2014, and 2018.

There are a few take-aways here. First, the probability of voting by age pretty steadily increases as cohorts age until about age 75 or so and then it drops off. Secondly, the slopes are fairly similar over time; it is a matter of levels for the differences in turn-out.  Young voters increased their voting rates faster than almost any one else in 2018.

Age is a strong dividing line. Pew shows this nicely from a poll in January 2019:

The oldest Millennials are just getting to the top of the first kink of the probability to vote curve.  Another year will make these old Millenials closer to forty and add another half a point or so increase in likelihood to vote during a midterm.  The eighteen year old kids who voted for the first time in 2016 will see a few more of their peers show up for the first time as twenty year olds.  Generation Z voters are just starting to vote in small numbers.  There will be far more new voters due to aging than lost voters due to mortality.

The core of the Trump minimum winning coalition is old people.  There is almost no juice left in mobilizing Silent Generation voters.  The youngest Silent Generation sliver will be at peak voter participation conditional on being alive.  Older Silents will be a declining share of the electorate for two reasons; there is a decrease in voting conditional on being alive to vote so some 2016 voters will be alive but not voting and there is significant mortality in this generation.  This generation will cast fewer voters.  Boomers are still slowly climbing the participation curve but half of the generation will be Medicare Eligible by election day so mortality will start fighting hard against increased participation.

Michael Bitzer has a good illustration of these effects on generation cohort composition in  North Carolina general elections.

The most Trump favorable generations (Boomers and older) are becoming less important as time goes forward.  Young voters came out in disproportionally large numbers in 2018.  The Presidential cycle will see Millenials and Generation Z vote share increase due to both the natural aging of the generation by another two years and the typical young voter boost in the general cycle as these are where the marginal voters are en masse.

It is a race, demographically, of whether or not the 18-40 year old population in 2020 can register and mobilize fast enough and decisively enough to deliver huge margins against Trump and for a single candidate or not.  And this explains some of the voter registration restrictions and hoop jumping.


37 replies
  1. 1
    Eolirin says:

    There’s a bit of a trap with looking at these numbers without the context of geographical distribution. While the overall picture the numbers present is heartening, the actual electoral consequences aren’t necessarily going to bear out until there’s an even larger drop off of older generations, depending on how those populations are distributed, especially for things like control of the Senate, and nevermind gerrymandered House districts

  2. 2
    Matt McIrvin says:

    I just got scolded on Twitter for pointing out that voters skew old, with people claiming that 2018 was a counterexample. 2018 seems to have brought youth voter participation up from obscenely low to slightly less obscenely low.

  3. 3
    Victor Matheson says:

    There’s also the statistical question about how much people change their political affiliation as they age. There is the old joke that if you are not liberal when you are young you have no heart. If you are not conservative when you are old you have no brain. (Of course that joke predates the modern brainecotomy of the current Republican party.)

    So, if people become even slightly more conservative as they age that demographic effect could easily swamp the benefits of more current liberals voting as they age.

  4. 4
    OzarkHillbilly says:

    @Victor Matheson: A # of studies have shown that political affiliations tend to stick for life. A young DEM voter becomes an old DEM voter. Same for GOP. YMMV.

  5. 5
    Matt McIrvin says:

    @Victor Matheson: I think there are *some* stage-of-life effects in addition to the cohort effects, but they’re complicated and not necessarily consistent between generations. I recall some old polling that indicated that GenXers have actually moved leftward as they aged out of young adulthood–as teenagers, they were more conservative than Boomers. But I can’t find it any more.

  6. 6
    sherparick1 says:

    @Victor Matheson: There is also the extreme irony that no one benefits more from Government programs than the us seniors between Social Security and Medicare. The plutocrats propagandizing that Government is about helping “those people” has never been more successful then in this one area of convincing a majority of white people over the last 40 years that Government is just about “taking” from them and “giving” to those people.

  7. 7
    tobie says:

    @sherparick1: If you live in rural America, there are two things you learn pretty quickly: Rural residents will always tell you that they were once poor and worked their asses off and never received a handout from anyone; and rural America lives off the generous aid directed its way from urban and suburban America because rural life is not financially self sustaining. The disconnect between these two is staggering.

  8. 8
    Catfishncod says:

    There is an implied equation here:

    Votes from any age cohort = Population x Citizenship x Registration x Turnout x Choice.

    (To get accurate results using higher math, each of the five terms should be treated as a function of age.)

  9. 9
    sherparick1 says:

    @sherparick1: A book I can recommend about what Trump, Mulvaney, the Heritage Foundation, and all their merry band of kleptocrats are doing to the rest of us, I can recommend Michael Lewis’s “The Fifth Risk”

  10. 10
    Matt McIrvin says:

    Here’s Pew’s results on party/ideological self-identification for different cohorts over time. It doesn’t go far back enough to show the shift I’m thinking of. It does show that polarization is increasing in all groups, but also the oldest groups that were already the furthest right are getting further to the right since about 2010. Fox News Geezers. Millennials seem to be moving further left as they get into middle age; with GenXers it’s kind of a wash.

  11. 11
    jonas says:

    @tobie: I also live in farm country and the way they square this is to say “well sure, we get subsidies, price supports, subsidized loans, etc. because [as you said] rural life is not financially self-sustaining, but we’re still busting our asses to help feed the country and the world…” implying of course, that *those people* get welfare, but *don’t* do anything to deserve it or earn it. The same logic applies to retirees when you tell them that they receive government welfare in the form of Medicare and SS: “Oh, that’s not welfare — I worked hard my whole life and paid into that system. I’m just getting my money back!” In a sense they have a point, but also fail to acknowledge the privilege of having been able to work steadily and earn those benefits.

  12. 12
    Matt McIrvin says:

    @jonas: Also, a lot of the actual work on these farms is done by farmhands who may be those people. Sarah Taber likes to talk about how the mythology of farming tends to elide the distinction between farmers and the actual farm workers, usually in ways that benefit people who are effectively farm management.

  13. 13
    Matt McIrvin says:

    …on the basis of just the Pew results, I’d say that if there is an age at which time makes you more conservative, it seems to be about retirement age. (Not “growing up and getting a real job and a significant investment portfolio”, which was the old stereotype.) And it could just as easily be the historical effect of the modern geezer-propagandizing network kicking in, because those groups were always further to the right of younger generations, over the range of polling at least.

  14. 14
    tobie says:

    @jonas: Like you, I don’t want to take away the subsidies that people receive. I just wish they would recognize how hard everyone works, especially poor people in cities who usually hold down multiple jobs and rely on an inadequate bus system to get here and there and have to deal with umpteen other hurdles just to get by. The subsidies I’m talking about don’t only pertain to farms. They include road repair, hospitals, emergency vehicles, electric lines, etc.

  15. 15
    p.a. says:

    @jonas: after x years they most assuredly are not working off their own cash. And in the case of major Medicare expenditures, x months.

  16. 16
    Belafon says:

    @jonas: The average adult male uses twice as much Medicare benefits as he pays in when he works.

  17. 17
    jonas says:

    @Matt McIrvin: That’s certainly true. I know guys, however, who still do most of the work themselves on their small family farms — usually also working a day job to pay the bills — and have the mucked-up boots, gnarled hands and bad backs to show for it. Around these parts, if you have even a couple of farm laborers on your payroll, you’re “a big operation.”

  18. 18
    swiftfox says:

    @sherparick1: For this post I’d skip to the chapter about the (former, of course) USDA rural programs administrator. Game, set, match.

  19. 19
    Villago Delenda Est says:

    Boomers suck. There, I said it, and I am one.

  20. 20
    Zach says:

    One of the key motivations for voter ID laws is to depress college student turnout. I registered to vote in my college town, and if I’d needed a state ID to do it I probably wouldn’t have bothered, and I wouldn’t have bothered to vote absentee where my parents lived, either… I was much more politically active/interested than the median college student. Also even for people who don’t change states, it’s more common for young people to change addresses, and updating IDs and voter registration can lag that by a long time.

    It’s not talked about as much, because there’s no legal or constitutional protection against age discrimination, but I wonder if the effect on youth turnout is as big or bigger than the effect on minority groups. It would be very interesting to see some targeted polling and/or modeling try to figure out what the impact is of these policies. Probably you could compare shifts in turnout in bordering states that did/didn’t implement voter ID in the midwest.

  21. 21
    The Very Reverend Crimson Fire of Compassion says:

    @Matt McIrvin: In much the same way that slave owners “produced” what came from the plantations. The tradition of pretending the white man on the porch works the fields is as old as the country.

  22. 22
    msdc says:

    @Matt McIrvin: That has more to do with differences within the generational cohort. In the 1980s, Gen Xers were conservative–because they came of age under Reagan. (Political identity tends to solidify around age 20.) In the 90s, younger Gen Xers (like me) voted for Clinton. Those patterns have held up over the years, with older Gen Xers (in what was at the time the 30-45 cohort) bailing Bush out in the 2000s and younger ones flocking to Obama. It’s not so much the cohort drifting leftward as it is differences in which part of the cohort falls within which of the typical age markers (18-30, 30-45, etc.) for any given election.

  23. 23
    patrick II says:

    @Victor Matheson:
    That has a lot to do with timelines. Younger are more interested in long term solutions. The most extreme example is climate change, where old people are going to die off before the beachfront reaches Atlanta, the young will have to face more of the consequences of severe environmental problems in their lifetime. And the young have a different set of issues, like anyone who still has school and family in front of them would.

  24. 24
    Immanentize says:

    I have a clear take away from these charts:
    Bernie is too old to be President. Biden too, almost. They are racing death just like their whole cohort

  25. 25
    Walker says:

    I disagree with the claim that this is a great graph. The colors for 2010 and 2018 are way too close to be distinguishable. So while the data may be great, the presentation is bad.

  26. 26
    Matt McIrvin says:

    @Zach: I’ve thought this doesn’t get enough attention for a long time. I think a significant part of the reason young people don’t vote is that right when they reach voting age and ought to be developing habits, they go off to college where there are often measures in place to intentionally keep them from voting, using their temporary addresses and possible out-of-state residency as a weapon against them. And most just don’t bother. It’s all very well to say they ought to have these presence of mind to jump through any necessary hoops, but whenever some people have special difficulties that others don’t, that’s going to skew the election results.

  27. 27
    Felanius Kootea says:

    @Matt McIrvin: Looks like the Democratic Party needs to do more to make voting easier for students – start with blue states where they control legislatures and then try to make inroads in red states. The expansion of automatic registration and vote by mail has already had a positive impact.

  28. 28
    Matt McIrvin says:

    @Felanius Kootea: New Hampshire and Maine, whenever Republicans have enough control, are notorious for trying to scare college students away from the polls with threats of prosecution for voter fraud if they don’t have in-state driver’s licenses (even if this is not what the law says). I remember during one episode of this in NH, there was a moment when one legislator flat-out said something like “these kids shouldn’t be voting at all, they don’t have the maturity to make good decisions”. Laying it all on the table.

  29. 29
    Ruckus says:

    Both will be in their late 70s – early 80s and if nothing else will have a rather different view than the majority of citizens. Especially those citizens not within 20 years of retirement.

  30. 30
    Raoul says:

    @OzarkHillbilly: I think this is more or less correct. Part of what McConnell et al (and the GOP states) are attempting to barricade against through voting restrictions is that party affiliation is pretty sticky.

    And the generation that grew up in Reagan’s shadow, rather than loving him and the GOP, loathe him, and Dubya Bush sealed the deal.

    Also not covered in the third graph is the demographic mix of the generations, which get more diverse as we move younger. While Hispanic voters might by 25-30% Republican, other POC voters, not so much. And even 30% is pretty devastating for a party’s long term success.

  31. 31
    Matt McIrvin says:

    @Raoul: My impression is that demographic mix explains most of the change. White millennials and Gen Zers are in some ways almost as reactionary as their parents. But proportionally fewer of those cohorts are white.

  32. 32
    New Deal democrat says:

    Two points:

    1. In response to Victor Matheson, no, people don’t tend to change viewpoints (become more conservative) as they age. People tend to form their core beliefs in their late teens and maintain that point of view forever. For example, the FDR generation voted reliably blue until they died.

    2. The demographics slice somewhat differently if you go by presidency when they turned 18. Mid-Boomers are reliably blue, while the youngest Boomers and early Millennials (turned 18 during the Carter or Reagan presidencies) are even redder than the Silents.

  33. 33
    Enhanced Voting Techniques says:

    I wonder if Trump’s appeal to his base is basically Trump gets to pretend he’s a teen until he dies? Remember the Boomers were the ones who weren’t get old “don’t trust anyone over thirty” and all that, and by now most Boomers have spent 65 years of being over thirty.

  34. 34
    Barbara says:

    @New Deal democrat:

    1. In response to Victor Matheson, no, people don’t tend to change viewpoints (become more conservative) as they age. People tend to form their core beliefs in their late teens and maintain that point of view forever. For example, the FDR generation voted reliably blue until they died.

    This is why Bill Clinton was elected — he received a high percentage of the tail end of voters who came of age during the Roosevelt administration. Indeed, that group voted for him by a larger margin than any other age group.

    One of my favorite websites:

  35. 35
    Matt McIrvin says:

    @New Deal democrat: The Pew results suggest that the Boomers and Silents actually did become much more deeply right-wing in their old age. They had been fairly right-wing all along (and less Democratic than their parents), but they got more intensely so during the Obama years. On the other hand, the minority who were liberal also became more liberal.

  36. 36
    Matt McIrvin says:

    @Immanentize: The Bernie fans argue that Bernie’s age works in his favor: his generation had real socialists, whereas the GenXers in the rest of the field who grew up under Reagan are too young to understand what that means.

  37. 37
    Barbara says:

    @Matt McIrvin: Wow, that is some serious stupid at work there.

Comments are closed.