Over the last several years people have been led to believe a number of ineluctable demographic truths, most of which turn out to be almost exactly wrong. (I wrote a book about this a few years back, which can loosely be summarized as: “Everything you think you know about demographics is wrong.”)
True to form, demographic trends continue to defy commonly held beliefs. For instance, it has long been an article of faith that the fertility rate of Hispanic-American women is so high that in the near term there will be a great shift in the racial makeup of America. But that shift has largely already been accomplished. Over the last 25 years the fertility rate of Hispanic women in America has fallen at a faster rate than any other demographic group.
The replacement fertility rate—the number of babies each woman needs to have in order to keep a population stable—is 2.1. In 1990, Hispanic women in America had a fertility rate of 2.96, far above the national average. By 2000 it had fallen to 2.73. By 2010 it was 2.35. The most recent numbers, from 2014, have it at 2.15, just a hair above replacement. Every other demographic group has seen small declines in fertility during this period. Hispanic fertility has dropped by almost 30 percent and is still falling.
There’s no telling where the floor will be on Hispanic fertility—maybe we’re already there, or maybe Hispanic-Americans will wind up with very low fertility rates, similar to Asian-Americans. I don’t know and neither do you.
My point is that both the hopes of the diversity mandarins the fears of the nativists have turned out to be incorrect. And that anyone looking at the actual numbers over the last 25 years would have seen that this was the direction in which the curve was sloping.
If that comes off as smug, I should lock it down. Because even people who follow demographics closely can be surprised. One of the big surprises over the last decade has come with the matter of “ideal fertility.” The “ideal fertility” rate is the number of kids that people say they’d like to have in a perfect world. Demographers and sociologists had long assumed that the average ideal rate of a country could never go below the replacement rate. Then, a few years back, some European countries started posting ideal rates far below replacement. And suddenly everyone started wondering if notions about ideal fertility were more subject to norming and re-anchoring than anyone had previously thought.
Another assumption, long held, relates to parental happiness. When you measure people’s “happiness” over time (this is a long, semi-confusing metric; you can either take my word for it here or go read the chapter in the book) what you see is that when you take two people identical in every respect—race, education, income, religiosity, geographic location, etc.—and one of them has a child, their happiness level drops. The parents lag the non-parents in happiness until the kids grow up and leave the house, at which point the parents start recovering their happiness while the non-parents tick backwards a bit in happiness. This is genuinely thought of as window-shopper’s remorse and the dictum in sociology is that, in the end, parents don’t regret having kids, but non-parents sometimes regret not having them.
And now there’s a lot of chatter about how this may not be true either, that women “regret” motherhood at a much higher rate than previously thought.
Unlike with ideal fertility, however, I find this revision somewhat unconvincing. It’s based in large part on a study by Israeli sociologist Orna Donath called Regretting Motherhood. But this isn’t a real “study”—not in the science-y sense of the word. It is instead, as the paper’s subtitle says, a “Sociopolitical analysis.” It was published in a journal called Signs, which is more like a feminist fanzine than a place for rigorous data analysis. And the crux of the “data” was that Donath found 28 women who said they regretted having kids and then interviewed them at length.
I don’t mean to imply that Donath’s work is useless—I’m sure there’s value in it on some level. Only that the journalists who wrote up her paper seem to think that she was presenting a new sociological finding rather than assembling lifestyle-justification anecdata. And these journalists were wrong.
Down the line, people who have children still seem happy with their decision. This is a truth. Just don’t ask me to explain it.
The final bit of revisionism comes in the world of transgenderism, where news reports breathlessly squealed that the estimated number of transgendered people in America is now 1.4 million, which is double what the best guess was just a year ago.
Before we go running off with this new number, a few caveats:
1) The 1.4 million number comes from the same place as the old estimate of 700,000: the Williams Institute, which is a transgender advocacy group.
2) When you find that your data is off by 100 percent after just five years, you ought to ask yourself some questions. Was the old data that wrong? Could the new data be that wrong? If both sets of data are correct, what is the dynamic in the population moving the numbers so quickly and how can you get any accurate measurement in the face of such currents? If neither set of data is correct, are your definitions too elastic to be useful? Or is the population too small to be accurately measured?
So far as I can tell, the response to the new “1.4 million transgenders” was not to ask any of those questions, but rather to cheer 1.4 million transgenders! Hurray!
3) When you look at the non-activist sourced numbers, they go in the other direction, with the upper-register guess being around 90,000.
4) What would you guess is the ratio of transgender-to-homosexual individuals in America? This is anecdata, but bear with me. You don’t have to say it out loud. Just keep it in your head, “From my life experience, I would say I know X number of transgendered folks for every Y number of gay or lesbian folks I’ve met.”
Okay. Got it?
The new 1.4 million trans number means that 0.6 percent of the adult population of America is transgendered.
The best numbers we have on the gay/lesbian population come from a massive CDC study in 2014 that found that 1.6 percent of adult Americans identify as gay or lesbian. So if the new trans numbers are correct, that would mean that there are only 2.7 homosexual people for every transgendered person in America.
Does that seem about right to you?
