Resolves to the half year (in local time) when @rachel and my child is born.
H1 refers to January through Jun; H2 is July through December.
Has to be a birth of a child we consider ours. Adoptions don't count; 50% parentage would if we get a surrogate mother or sperm donor for a child that we raise.
Resolves to "2027 or later" if we still don't have a child by the end of 2026.
We're currently thinking about having our first child after a couple years of being married, balancing the desire for Rachel to settle into her career vs the desire to have 4+ kids and the uncertainty about AI timelines.
Related questions
@traders Rachel wrote up her thoughts on "choosing to have kids young" on her new Substack: https://rachelweinberg.substack.com/p/why-baby-now
@Joshua yes that's the due date haha—though there is still some alpha to be gained if one puts time into weighing the following factors (which I haven't done):
- more babies are born early than late
But...
- multiples basically guarantee an early birth, so that takes up some of the early probability, but we have a singleton
- I really don't want to have a c-section, and planned c-sections also account for a bunch of the early births. Same with induction.
- I'm really young and have no known risk-factors so am less likely to have complications that cause them to take the baby out early (e.g. gestational diabetes)
- there's still a ~2% chance of miscarriage, which would likely result in us having a baby H2 2024 or H1 2025
Maybe some other things also.
Once you're convinced it's a good idea to have a child at some point, your bias should almost always be towards sooner rather than later. There are plenty of valid-seeming reasons to delay ("I'm not ready! I can't afford it!") but people alive today in Western countries are, in almost all respects, better-prepared than 99%+ of their ancestors for having kids.
The one area in which we're not better-prepared is access to alloparents, whether through friends/community or relatives. This is outside of the historical norm. However, one way to get access to alloparent care over time is to have many peers who are also parents; child supervision scales more than linearly, so when multiple families have similar-aged kids, a subset of the parents can watch the kids while another subset does something else (once a month, a bunch of the dads in my friends group watch all the kids while the moms escape and hang out together. It works well, because at some critical mass the kids can all entertain each other with minimal adult supervision). And how can one encourage this? By having normalizing reproduction through having a kid.
Another reason people in tech in particular should increment their target-kid count and reduce their target-first-kid date: QSBS! https://actecfoundation.org/podcasts/2021-qsbs-qualified-small-business-stock/ If you plan to make angel investments and expect to be successful at it, each additional kid gives you the opportunity to save a ton in taxes later on. Under the right assumptions, kids can actually be positive expected value! (Be sure to run this reproductive decision by a CPA, of course.)
In conclusion, have kids early and often. Even if you're unprepared right now, you have nine months to fix that. A baby gestates over 40 weeks, boot camp for the marines is 13 weeks, obviously this timeline gives you plenty of room for transformative personal experiences even if you're also busy with a day job.
@ByrneHobart I mostly agree with your comment, but I think the law of equal and opposite advice applies here. We're already talking about having kids very early relative to our social circle/professional peers: I'm 20, and planning on having kids around age 22-24. And I think more people should do this, so for most people your comment is directionally correct! But I haven't graduated college or worked full-time for a year yet. I actually do think I could take care of a baby fine now, I'm just worried about it taking away time from work, which it clearly does, especially for women (30s is when the wage gap widens a lot). I'm happy to make that sacrifice, but I think it makes sense to invest in my career a bit more first.
@rachel of course a lot of it depends on goals and preferences. It definitely takes time away from work (my wife worked full-time for a bit before the first baby, part-time for a long time after, and switched to full-time SAHM status ahead of #4, who is due in July). That did lead to an increase in the household-level wage gap, but left everyone better-off; some of the wage gap is just a measure of the returns to specialization (this is one reason the intra-family gender wage gap is much starker among high earners). The second income + paid childcare combo is pretty disfavored from a tax standpoint; the incremental income is taxed at your marginal bracket so the conversion from pretax dollars to after-tax childcare is pretty poor (and this is even more punitive in high-income cities where a) local taxes are a lot higher, and b) the cost of childcare is extremely high). In a sense people are broadly correct when they say that there's a tradeoff between kids and career, they're just taking the wrong side of the trade. So aside from QSBS favoring more kids, state and federal income taxes are a subsidy for kids first / career second.
Investing in the career is a good way to think about it, but one way to extend that is to note that the assets you're getting—skills plus network—will tend to depreciate when not in use. So if you're paying the market price in time for those, but planning to utilize them less than the average person, you're getting a suboptimal deal. It may work out for other reasons, but kids-then-work allocates personal capital better than work-kids-work. (And I tend to find myself saying "this would be easier if I were ten years younger" more often on evenings and weekends with the kids than during the day, though different careers index to skills that are age-sensitive in different ways.)
Additionally, there's the positive externality of doing something weird: we have a lot of information on what it's like to defer childbearing in order to enter the workforce, much, much less information on what it's like to do the opposite (at least among the "uses prediction markets and links to SSC" demo). Society is stuck at a local maximum in the great multi-armed bandit experiment on optimal life paths, and highly divergent choices can provide enough information to either confirm (">it's a good maximum, sir") or not (">it's local").
@ByrneHobart Oh the tax stuff is interesting, I hadn't thought about that angle before. Though ultimately the finance stuff isn't a huge factor for me here.
The second point seems very true—I remember complaining a few years ago that the structure for young women was pretty suboptimal and the opposite order would be better (for fertility/energy reasons as well), and this is actually doable for me since Austin is older so we have enough money to have kids now.
But I'm pretty focused on AI and to the extent I care about having a career it's mostly to make sure all of that goes well. Based on my timelines + the number of kids I want to have, career after kids isn't really an option: I expect by the time I'm 40 the important work will be done. Or we'll be dead. Besides, I plan to work when I have kids, just more like a flat 40 hours rather than the 50 or 60 that I can do pre-kids, so my network and skills will stay in use.
Also agree on the last point. This is tempting to me, and part of what's exciting about having kids at 22-24, which is still quite abnormal.
@rachel OK, conceded; a first-principles argument for having kids will have a hard time beating a first-principles argument for working on AI right now. As to whether it should, that's a question that future generations of kids or AIs will have to answer for all of us.
@ByrneHobart re: positive externality -- I can't speak for Rachel, but I'm very open to negotiation if some eccentric pronatalist billionaire wanted to subsidize the externality eg by offering $50k for us to have a kid in 2024 (aka about 1 year sooner than the market currently says) and blog about the experience.
Also, relevant tweet:
I'm not sure how to answer this question candidly because we were already planning on having an amorphous largish amount (4+) but I could imagine "$1m brings us from 5 to 6"
@barak Resolves to "2027 or later" if we don't have a child together by the end of 2026. (Thanks for asking, I'll update the description)