What Keynes Can Teach Us About Trans Healthcare

I promise this will make sense by the end

The difference between risk and uncertainty runs throughout John Maynard Keynes’ work, especially in “A Treatise on Probability”. The basic idea is this: If you're just operating under risk, you know the full range of probability outcomes, and can make decisions accordingly. Under uncertainty, there are no such guarantees. You have some information, and attach some level of belief to it, but that's about it. Yet as Alex Williams describes, we can still make decisions:

In real life, uncertainty comes from many places. It comes from data itself: selection effects, measurement error, etc. As mentioned, there’s also meta-uncertainty about the models we use to analyze the world. If we are fundamentally uncertain about which model is correct, how do we choose which model to use? Williams suggests a heuristic here: we should determine how much damage, especially irreversible damage, is caused by picking a model that ends up being wrong. The bar is simply higher for models that may cause such damage.

---

There’s been a lot of articles recently about transgender children, especially what healthcare is provided for them. A recent study tracked feelings of anxiety and depression among young folks before and after 2 years of gender-affirming hormones (GAH), and found those feelings improved somewhat. Some authors pick this kind of study apart, saying those feelings might have improved even without GAH. And sure, having a clean methodology on this is hard. Trans people are a small subset of the population, so sample sizes are often small. Many studies over a longer time period will lose contact with a big chunk of participants before the studies end. There will be physical changes from GAH (that’s kind of the whole point), so you can’t do a double-blind study. And beyond the ethical problems, just denying people GAH to try and get a control group means many folks will just get it through other means.

One problem with these articles is that they deeply scrutinize studies that show good results for gender-affirming care, but don’t apply the same scrutiny to studies or anecdotal evidence on the other side. And yes, there is some solid data we can point to, like this meta-study showing regret rates for gender-affirming surgery are less than 1%. But I think the problem is deeper here: these writers just simply have a bad model for thinking about trans people. These articles fearmonger by repeating the idea that we “just don’t know” how these things are affecting kids. This is solidly in the realm of uncertainty, then. We’re not discussing the exact percentage of who will detransition or see mental health improvements or whatever. This is solidly in the realm of decision-making under uncertainty then, and we need to be careful to pick the right model.

Today, we will be discussing a writer who was not careful.

---

Recently, Matt Yglesias published an article entitled “Progressives need to engage with the specific questions about youth gender care”. His thesis is this: abstract values about supporting the dignity of trans people, “the guiding principles that help us push back against their conservative crusade”, don’t help guide us on factual questions of when certain medical treatments are appropriate. Let’s be precise about what this value means: affirming that trans people are not lesser than cisgender people. That their gender is just as valid, that they deserve respect and not vilification. This is really a model, a heuristic for how to think about trans people. And this model is a safe bet, given the alternative here is to support the rhetoric of genocidal fascists. I think these values serve as a good guide for what Yglesias brings up.

First, he makes a comparison to the opioid crisis. He says “The problem with the whole thing wasn’t the abstract values Purdue Pharmaceutical was appealing to — it was in the specific details of the implementation.” But this comparison has some problems. “Not worrying about addictiveness of painkiller” is certainly a bad value that caused problems here. And we respect trans people’s beliefs about the bodies they want: not so for painkillers.

Yglesias goes on to cite a few New York Times comments, saying “the structural commonality among their points is that affirmation of an equal right to human dignity does not determine a unique answer to all of these questions.” But the first comment is just denying the validity of trans kids because there are a few who live near someone. The second comment¹ is very insightful: the person first says “Biological males competing in women’s sports is a problem, at least after Middle School”. Presumably, that’s because puberty introduces some permanent physical changes to kids’ bodies. They go on to express their concern about trans kids consenting to “medical treatment that is in many cases irreversible”. We'll get back to that in a second. The last comment has a woman saying “the difference between male and female humans need to be respected” and that because “women's rights are in no way complete or guaranteed… we need our spaces.” Our spaces? If “our” refers to women, then obviously trans women should be included in those spaces. Sure, women face unique oppression that requires some unique spaces: but that’s about being a women, not being cis.²

Yglesias next cites some trans folks saying that kids are too young to transition if it causes irreversible changes. He boldly proclaims “This is where you get into the reality that abstract political values don’t always answer factual questions.” Let’s take a quick aside, since this line about “irreversible changes” is so common.

---

Per our discussion of Keynes earlier, harming people in an irreversible way is a concern! But Yglesias is making a simple logic mistake here. Twitter user @bimbotheory put it best:

cosima bee concordia (Villain Arc) @bimbotheory

As much as you may hate to admit it, denying trans kids care and forcing cis kids to transition are morally equal if you actually believe transness isn’t inferior to cisness 🥰

I really feel the need to emphasize that if GAH are not fully reversible, normal puberty isn’t reversible either. If kids consenting to going on hormones is problematic, them consenting to go through unassisted puberty is equally problematic. These applies to Yglesias discussing puberty blockers too: maybe it’s unclear if they have irreversible effects. But those effects do seem minor compared to the irreversible effects of puberty.

Going through the wrong puberty is certainly harmful and has some permanent effects. But do you think the number of cis kids incorrectly being put on hormones is anywhere close to the number of trans people going through the wrong puberty because of a lack of healthcare? Some people on the margin may be incorrectly given hormones, but LOTS of trans people are obviously incorrectly denied care.

Yglesias really starts to give the game away at the end. After an aside about some doctors who are giving care to trans patients more quickly³, and a reference to an uncorroborated "whistleblower", he has a very interesting passage:

There’s an internet trope about the guy who’s “just asking questions.”

You don’t want to be that guy. The “just asking questions” guy is operating in bad faith or with a desire to wound people or be cruel or impugn their dignity.

…

For all the reasons Bouie outlines, it would be in many ways convenient to stand in unequivocal solidarity with people whose dignity really is under assault from a conservative movement that hasn’t reconciled itself to the gay rights revolution but knows it’s too politically weak to overturn it.

So what reason does he give for not standing in solidarity? Well, because a lot more kids say they are trans, and “the increase is preponderantly among people with XX chromosomes.” These may be interesting questions to study⁴, but it's very easy to say "trans identification is changing, and trans people are still valid". Yglesias goes on to imply that people are "pathologizing girls and young women who don’t conform to patriarchal definitions of femininity" and that "we are now in a new era of medicalizing teen girls’ discomfort with patriarchy while downplaying what appears to be a widespread youth mental health crisis". His implication is clear: we shouldn't assume trans kids are valid. Instead, they are suspect, confused, and mentally ill. Does this sound like supporting the dignity of trans people?

The structural commonality among Yglesias’ points is that affirmation of an equal right to human dignity does, in fact, determine a unique answer to all of his questions. I only wish he could see that.

It’s notable that Yglesias did not cite the regret rates mentioned earlier. Some might contest that those studies were done awhile ago, and the new standard of care for trans kids is different, or that we don’t know the long-term effects of these treatments. But returning to Keynes, the question is this: will we wait for total certainty about trans healthcare? Or will we act to protect the trans children under attack right now across the world?

1. This comment also brings up the “trans women in women’s sports” complaint, which was always a weird one to me. The core premise seems to be that trans women, perhaps by nature of their birth or because they went through male puberty first, have an inherent advantage. It isn’t necessarily a problem in theory : men who are naturally much taller and twice as big as me competed right next to me. Having genetics that makes you better at a sport is not disqualifying! And it certainly isn’t a problem in practice: the go to example here seems to be Lia Thomas, a trans woman who is ranked… 46th nationally. Seems fine!

2. I’ll get ahead of potential objections here by noting that sure, there are some general differences between cis and trans women. But these people want to exclude trans people for being trans, not for any actual feature. Lots of trans women have vaginas, lots of cis women don’t have a uterus, and so on.

3. If you were really concerned about a lack of evidence here, you would want people to be trying out a wide range of treatment approaches. Plenty of examples of accessing care being nearly impossible, let’s get some data on how making it really easy goes!

4. The point about rates of trans identification isn’t convincing: society is getting less transphobic, we should expect more trans people. Meanwhile, the Williams Institute study on the matter notes this:

A recent study estimated that nearly one-third of transgender adults identify as nonbinary, which is similar to our finding of 25.6%. The YRBS does not include a follow-up question to allow respondents to further describe their gender identity. Therefore, we are unable to provide a more detailed description of gender identities among youth.

It’s very plausible to me that a lot of the increase, especially among girls, would be non-binary kids who won’t transition and are not relevant to this conversation.

This data from another linked study about an increase in AFABs being trans is… not great. You can’t seriously look at this sample size, with this trend and that 2013 outlier, and tell me I should be worried.