The need for a structured framework for deciding whether you should have children
In the rationality community and in the EA community, it is normal to analyze all kinds of decisions in detail, and this of course includes their ethical consequences—that is, the impact of decisions on other people. Yet even though there are a number of forum posts on the decision whether to have children, they often focus on the effects on personal productivity and resources or personal happiness of the potential parents. There are also some ethical considerations, but the effect of being brought into this world on the child itself receives surprisingly little attention. In general, there does not seem to be a place where all the factors of the “should you have children” (SYHC) question are brought together in structured way. This can lead to implicit normative assumptions and vagues ideas heavily influencing discussions.
Deciding whether to have children is one of the most important decisions in your life as it will require a large amount of your time and financial resources and will likely be a life-changing experience that will transform your worldview and preferences.[1] Of course, not having children can also drain your resources because, depending on your personality and values, you might feel incomplete and unhappy.
With this post, I aim to organize the SYHC discussions and offer a framework for the considerations that are necessary for them, in order to enable a more structured and rational discussion for those who are interested in the topic.
First, I discuss how to approach the question as a rational decision, and discuss the considerations that should, arguably, be part of a rational decision on whether to have kids. I split this into considerations regarding the parents, the child, and the world, discuss how these considerations relate and which role time horizons, discounting and timing play. After an interim summary, I discuss failure modes of rational discourse and rational decisionmaking in this context, with one particular fallacy that I suspect plays an important role in the SYHC decision. I also mention some alternatives to having children.
My conclusions do not include an object-level answer to the question whether you should have kids. In particular, this post is not intended to say that you should not have kids or that you can easily sum costs and benefits for this decision. I’d welcome discussions, including about empirical insights into the costs and benefits, in the comments to this post.
The decision of whether to have children as a rational decision
Suppose you contemplate one of the most important decisions of your life, namely, whether to have children. Like any other important decision, it will have positive and negative effects—costs and benefits -, and it comes along with uncertainty, possibly including uncertainty about your own preferences. Considering all these trade-offs can be challenging, and it can be even more challenging to reduce the uncertainties given that many people have very strong opinions about the topic (and may understand your questions as an attack on their decisions or values), nearly everybody has implicit (ethical or empirical) assumptions, and many other people may not find the aspects even discussable that you care about. However, as with other decisions, if you are really unsure about how to decide, it can be helpful to actually think about the factors that influence your decision. If you are not unsure, thinking about these decision factors can help other people who are unsure.
The simplest decision-making framework is: ”Do something if the expected net-benefit of doing it are positive, where net-benefit is benefit minus costs.”
I’ll ignore the uncertainty for the moment and just write for the decision criterion.
For many people, this sounds fine for business investments, but awfully cold and calculating in the case of children. However, if you consider “benefits” as positive deviations from the situation you would be in if you had no children and “costs” as negative deviations from that situation, it is a useful framework. Of course it can include the effects of having children on the child and on rest of the world.
where and are weights that depend on your preferences—that is, shows how much the child’s expected net-benefit influences the decision.
As a rational decision framework, this has limits, but it makes sense to first see how far we can get with the framework.
seems strange—after all, isn’t all of the of the parent? Yes, but there are more direct costs and benefits to the parents. This includes, for example, pregnancy sickness.
Note that what counts as costs and benefits may be very different from person to person, depending on your preferences, your biology or your opportunity costs. Therefore I will not list costs and benefits strictly separately, because many factors can be costs or benefits depending on your preferences. (For example, spending time watching cartoons with children can be fun for some people and annoying for others.)
For many of the factors discussed below, your personal preferences will matter a lot, and obviously your expectations about your personal situation and your life’s development are important. However, with respect to your (potential) child’s life and your child’s effects on the world, global developments are very important as well.
Social pressure, approval or disapproval can be part of the . This is also an example that you can influence some costs or benefits by changing your environment or having different (maybe more stoic?) attitude about them. At the same time, it is an example that there may be costs that you can only shape but not avoid—if you feel social pressure that you should have a child, then keep in mind that you will also feel social pressure with respect to your parenting decisions once you have a child.
The framework does not imply or require that you make this decision alone. Maybe you and your partner are thinking about these issues, maybe you have completely different opinions, for example about priorities, preferences, expectations.
It is important to remember that all of the factors influencing your decision are uncertain. You can invest time in personal research to become more certain about them, but the best you can get is subjective probabilities.
With all this in mind, let us consider the different categories in more detail. The lists are open for further additions, though they also try to include everything that has ever been discussed in this forum; see the survey “Should you have children? All LessWrong posts about the topic”.[2]
The costs and benefits for the parents
These are points to consider from the parents’ perspective:
How much do you like children? Do you like “the feeling of a soft warm baby in your arms”, spending time with children, playing with children? How much would you enjoy being parents? How much joy do you expect to get from playing with the kids, experiencing their growth, watching them be kids and grow up? How much would the less pleasant tasks of parenting burden you?
How much do you expect the same to be true for becoming grandparents when (and if) your children have children?
How do you value the time you will (have to) spend with your children compared to other ways to spend your time? Children are very time-intensive and limit your autonomy with respect to your own time. They are a long-run obligation and thus a strong commitment.
Note that this does not constitute a cost with respect to the time that you actually want to spend time with your children, but it makes sense to think of this as “benefit = time spent with my children”, “cost = forgone trek crossing the Alps”. Would you travel the world if you had no kids? Would you spend your days partying? Would you save people from starvation? Would you solve AI alignment? This is a more important point for people who have very strong wishes how they would like to spend their time, or feel obliged to spend their time in certain ways.
However, note that time usage is not always rational. Instead of preparing packed lunches for your children, you might just binge-watch a series on TV and regret it later.
The amount of time required may depend on whether you have support, such as your own parents living in the same village, or if there are good kindergartens.
You may also believe that children are even more time-intensive or costly than it would reasonable, due to social pressure or wrong assumptions about your effects on the child’s development.
Raising children is financially costly.
What would you spend your money for if you had no children? The cosniderations are similar to those with respect to time.
This argument is sometimes made in particular noting that you could spend the money for altruistic causes. However, it also applies to people who just want to spend the money for themselves. Consider however you would spend the money if you had no children.
You much do you expect exhaustion / loss of sleep to bother you or to affect your health? (I don’t want to overstate this—people are surprisingly adaptable, and you can optimize this, like many other things. But it’s worth doing your own research here.)
How much do you expect having children will affect your health? (Children spread pathogens. But maybe you are more active because you have children, or your children may care for you when you are old which is good for health then.)
Monetary and time costs are not completely separable because you can, to some extent, convert one into the other. The same is true for e.g. time and health/sleep or money and health/sleep.
Do you expect having children to increase or decrease your productivity?
Maybe you will fulfill a fundamental desire that would otherwise haunt you and keep you from working productively.
Maybe the time scarcity makes you prioritize better.
Maybe your motivation (to create a good world for your children or earn money for them or whatever) increases.
Or maybe you are more distracted and no longer able to focus.
Do you expect having children to increase or decrease your short-run happiness / life satisfaction for some other (maybe unclear) reasons?
Do you expect having children to increase or decrease your long-run happiness / life satisfaction for some reasons (e.g., expecting pride or companionship)?
How much do you expect to enjoy “creating” a new life, or expect that you enjoy nurturing, educating, parenting their children? How important is it to you that there will be beings that are relatively similar to themselves, or whose personality you at least strongly influence? Do you expect to feel emotional satisfaction or a feeling of meaning from creating a life, seeing the children grow, being responsible for their success? Do you expect that this gives you the feeling of doing something “real” that you otherwise would not have?
Does your life partner does or does not want to have children, and you want to do something that he/she wants?
Do you think that by having a child you can avoid societal or peer pressure? Or do you expect additional disapproval if you have a child, or strong prescriptions by the government, society or your peers on how to raise your children?
Do you expect to feel insecure/anxious? Do you expect to feel unable to raise kids? (If so, you may want to think more generally about how to reduce your anxiety. Or it might be helpful to know that you can get advice on raising children and that you will grow with the task.)
Do you fear status competitions that come along with stress and even more costs including more time costs? Note that all time and cost considerations, but also your stress level, is influenced by social expectations, in particular if you do not consciously decide how much to care about them.
Many benefits and costs (in particular, time cost, monetary cost, cost in terms of energy, health, but also feeling insecure or anxious) may be influenceable to some extent by planning, optimizing, coordinating with like-minded people, just talking to people, timing (e.g. having children at a time of your life when your own parents may be able to help because they retired but are still healthy).
The costs and benefits for the child
One topic that I don’t think is discussed enough is the impact of being born and brought into life on the child itself—the child’s perspective. Do you think that children born today can expect a good life today and in the future if they are raised by people like you?
The general quality of life depends on many things like the child’s family (including yourself), your income, the country you live in, the child’s biology and health etc.
Arguably, most people like to be alive. That most people don’t commit suicide is sometimes mentioned as evidence (though it is not clear to me how strong this evidence really is). Nonetheless, it seems plausible to me that most people like to live. Moreover, it seems that with today’s technology and resources, life could be very beautiful for everyone if manking coordinated better. Life can be improved on both the micro and macro levels. If you are reading this text, then you are a possibly a kind of person that will be able to read texts on how to do this.
It is sometimes added that people are better off now than they were in former times when people also had children, even in times like the Thirty Years’ War. However, this does not add anything to the argument. People decide differently, maybe they just do what everybody does, maybe they believe God commands them to have children. That you exist does not mean that your ancestors had a good life or even that they made good decisions. This confuses the “is” and the “ought”. Similarly, it does not add much that people in miserable situations today also have kids. It is, at best, a statement about the parents’ decisions. (Maybe and alternatively, the statement is a way of saying that contributing to the future of humanity is really important, but then people should say that more clearly.)
Is it good to be born even if you expect the world to end in a catastrophe in a decade? The question of “How much do you expect that your children will have happy and fulfilled lives?” would be strongly influenced by a catastrophic-risk expectation. The question is linked to the question of general expectations of humanity’s future. Thus, discussing whether the world will be worthwhile for children to live in should help to clarify what to expect from the future for yourself and how to prepare for it. Strangely, the relationship between expectations about Global Catastrophic Risks (GCR) like nuclear war, pandemics, and unaligned artificial superhuman intelligence[3] and expectations about a child’s life quality are rarely discussed even in a community that assumes that GCR have relevantly high probabilities.[4] It seems to be very important what kind of catastrophe the parents expect. For instance, if they expect an “everybody suddenly gets switched off by Nanobots” scenario, they may think that a 5-year life is positive, but it is possible to imagine scenarios that include much more suffering.
There are people who assume that living is bad in general. (For a rejecting discussion of this position, see for example “Antiantinatalism” by Jacob Falkovich.)
There are also people who assume that the future will not be worthwhile to live in; I have not really seen very good and credible discussions of expectable future life quality (with or without transformative AGI).
However, it is clearly also true that no life is only and always pleasant. At the same time, there is no reason to assume that is a binary. If you know that Utopia will soon start (whatever you imagine Utopia to) , then life will be positive on average; if you expect a dystopian S-risk scenario that lasts forever, then life will be probably something to avoid.
The costs and benefits for the world
These are points that affect the world—the “world’s perspective”, maybe the humanity’s perspective or the perspective of sentient beings.
Do you think that the existence of (additional) people is good for the world? If you think that the existence of people is good then you may contribute to their future by having children.
Arguments why the existence of people is good for the world may include that younger people are necessary to care for old people, or that young people are good for the economy. Do you think that people have positive or negative externalities? Is there strong resource scarcity for humanity, or do additional children add so much to technological progress that their marginal contribution is positive?
Do you think it would be good for the world if people like you or you in particular should have children (selective pro-natalism)?
Do you think that your children would make the world a better place?
Do you assume that it is better if “wealthy, educated and responsible people have more kids”? (This argument holds for wealthy, educated and responsible people—however, you may like other characteristics of yourself that may support similar arguments).
Do you want to contribute to the survival and spread of certain kind of preferences/memes/culture? (There are people who also see children as an “investment in AI research”)
Altruistic actions you give up because you have too little time or money are not part of because they are no direct effects of your child; instead they are part of , because the child changes your time or money usage.
It seems possible that the quasi-altruistic motivation of contributing to the future of humanity by having children may increase due to a catastrophic-event expectation, because maybe the number of people surviving the event is higher if there are more people before the event. But this depends on the kind of event. The “more people will be left” thought would definitely apply if some catastrophe is expected to kill a fixed absolute number of people or even a fixed share of the world population, but would not apply if for some reason the number of people expected to survive is fixed or it is clear that only people living in a specific unreachable area of the world will survive.
Relations between the three groups of costs and benefits
Many positive feelings for the parent () would probably vanish if they expected , so treating them as separable has limits. To some extent that may even be true if parents expected .
Even very direct costs may depend on the meaning attached to them. Spending your time taking your child to a guitar teacher and waiting until the lesson is over may be costly, but it would be much more costly if you did it for some stranger’s child, whereas you may not find it costly at all if you attach meaning to it, like contributing to your child’s ability to enjoy music in the future. Similarly, you may find the sleep loss less costly if you think that your child will have a happy future and also contributes to the benefit of the world.
Parental feelings that they have a “duty” of having children or that this is part of a “fulfilled life” may be part of but that only makes sense if .
Assuming that having children is good for the future of humanity () depends on the assumption that life is good in general (i.e. not for your child in particular, but for children on average, at least for sufficiently many generations in the future).
Note that all considerations depend on your time horizon.
Expecting a long life means that you can expect a greater , because some of your benefits only accrue when you are older.
If you expect your child to live longer, then will likely be larger, given that it is positive at all. Though some argue that happiness is highest for people when they are children and decreasing over a person’s lifetime, additional years add value if the for these years remains positive.
If you expect humanity to continue to exist for longer and its existence to remain positive, then investing effort now and contributing to the future development of humanity is more valuable. Expecting a high probability of an existential catastrophe for humanity will not only reduce , but also and .
However, if you think that humanity is about to get destroyed by a catastrophe and you can do something against it, then this decreases (because it increases your opportunity cost).
If you strongly discount the future, all of this may matter less—but this is also true for anything positive that would happen in the future (including later in your own life). This would justify having children if you really just want the feeling of having children right now.
Notice that a five-year time horizon (with an end-of-the-world endpoint) does not necessarily increase your opportunity cost for these five years. Sure, if your life goal was to visit the Louvre and you expect the world to end in five years, then you need to visit the Louvre now and not when your children are 18. However, if your plan was to build a big business, that will seem much more difficult in five years than it will in 20 years (and what’s the point of the goal then anyway), so you may not stick with that goal anymore.
In addition, your opportunity cost over the five years depends on whether or not you believe you can influence the likelihood and magnitude of the disaster. Some people seem to believe that having children greatly increases their focus or motivation, which may offset the necessary time investment.
Relatedly, costs and benefits may depend on timing and the order in which you do things in your life.
In general, you need a partner and biologically you can only have children during a certain phase of your life. The probability to be able to have children decreases when you get older. You may be too optimistic when you are young.
If you have children earlier, you will have more time overlap with them, which seems good for you and for the children (and for your potential grandchildren).
If you have children later in life, you may be more financially secure and have more life experience.
Some people argue that one can do other things first because in the (sufficiently near) future a superintelligent AI will solve all problems by either making humans immortal (which would then extend all time horizons enormously) or at least by overcoming fertility problems (including retroactively restoring fertility). This argument can of course also be made in a weaker form: perhaps you want to wait instead of having children now because you hope for technological changes that will improve the lives of your children.
Summing up the costs and benefits
Considering the costs and benefits listed above (and adding costs and benefits that are not yet included in the lists), it is possible to think about the SYHC decision in a structured way. Literally adding and subtracting will not work, because you would have to attach some monetary or similar value to the costs and benefits and that will often not work or not feel right. However, you can use such lists as a help to understand the consequences of your decision.In theory, you could write down your expectations for your life with children and your life without children, then compare these options and see the differences in each area. Conversely, ignoring costs and benefits does not mean that they do not exist.
You can also use this framework to discuss your decision with other people. If your partner has a different opinion, but also thinks that discussions are a good approach to decisions, then this may be a good starting point.
Even if your decision is already set in stone because according to your preferences, one of the points above is more important than all other points together, then you can use this framework to understand its consequences in other areas or understand people you disagree with.
Failure Modes of Rational Discourse
It is possible that rational discourse fails. The issue at hand seems particularly vulnerable to a failure of rational discourse. Some reasons for this:
People can feel very strongly about the topic and therefore not be willing to engage in rational discourse at all.
People sometimes pathologize other people, for example they claim that their worries must exclusively come from emotional distress instead of discussing their worries.
Parenting decisions are usually decisions that are made with your partner, so this requires communication. This kind of communication may be difficult because so much depends on it.
Some say that “whether to have children isn’t each other’s business and pressure against doing normal human things like this is net socially harmful” which may also lead to misunderstandings or fallacies. So be aware that some people may perceive discussing these issues as intrusive.
Some people think that discussing reasons and arguments on the topic is nonsensical. For instance, someone may think: “At some point in your life, you just have kids. It’s just what you do. That’s the way life is!” This kind of explanation seems similar to the “fulfilled life” or “duty” arguments already mentioned, but it explicitly avoids considering the arguments. This attitude may seem incredibly rigid, but at the same time, it may be very comfortable. If this is how you approach decisions, you do not have to weigh costs and benefits, and have you one thing less to think about. Most importantly, you will never feel responsible for any consequences. So even though this way of deciding may not be rational, it may still be good for some people. However, these people may not have a way to find out whether this is the case, and they may very well change their mind later on and regret their decisions (this of course may also be the case for people who approached their decisions in a more rational way).
“If someone tells you not to have kids then that is crazy and you should not even consider arguments about it. The voluntary human extinction movement argues for not having kids and that is obviously crazy.” Well, maybe there are people who do not know and understand your position yet? If so, then you can do them a favor by explaining to them why telling people not to have kids is crazy.
To discuss the considerations that influence the having-kids decisions, I included examples like GCR and suffering risk. There are more mundane fears like “Will I put my kids into a world that completely lacks meaning and perspective?” which may seem even more crazy—or at least that is the impression of answers I saw in discussion of such questions. I still could not explain to myself why these worries are not justified. Even a “positive” AI future, as little as I can imagine it, seems worrying to me. However, I tell myself that this is just a typical human reaction—many people reject any kind of change.
Failure Modes of Rational Decisionmaking
An even more basic failure mode is that of rational decisionmaking. Why may that failure mode be relevant here?
Emotions (urges, desires, feeling insecure) can be very powerful. Hormones etc. can be very strong, in particular during some years of your life. This can give rise to rationalization.
Having children may completely change your preferences, so it can be a transformative experience. The decision may feel very different after you made it.
We do not completely know our own preferences. Therefore, it makes sense to consult statistics on happiness / life-satisfaction; but people question whether happiness is actually the same as long-run preference satisfaction.
Happiness research seems messed up. Are people nowadays really as much happier than 100 years ago as you would expect based on your knowledge about economic growth? Also, how do you really evaluate a life’s happiness given the “Two selves” that Kahneman explains in “Thinking, Fast and Slow?
Time-Inconsistency: Another way that decisions are irrational is time-inconsistency. Consider someone who describes his or her decision like this: At some point in your life, you feel overwhelmed by some vague but extremely strong desire to have kids. That is because you just want it, or because of your hormones, or whatever kind of perspective or level of explanation you prefer. This suddenly overrides all of your previous plans for your life. If preferences are like this, they are immune to rational considerations. This is similar to the desire of a drug addict to get more heroine; you can tell him that drugs are, all things considered, bad, so he should be happy that there are none available; but he may not be happy. Actually, there is a very specific way your preferences have to be over your lifetime that does not give rise to time inconsistency (adding up utility for every moment of your life, exponentially discounted or not discounted at all), and it is not clear that brains work that way. Making consistent decisions over time is hard and nobody prescribes what your preferences should be.
Regret and alternative realities: Which kind of regret disutility do you fear most? Do you fear that people say “I wish I had never been born!” (probably not, you can always say “I did my best and all other people had kids) or do you fear that on your deathbed you think “I wish I had had kids”? In the latter case, on your deathbed, you might compare murky reality to an optimal hypothetical life and therefore feel a lot of regret. This motive of avoiding regret may not be compatible with rational considerations because the object of regret is itself not grounded in reality (or in realistic expectations over what could have been).
Confirmation bias: People may stop taking in new information once they have made a decision, because this information may question their decision, which creates a bad self-image.
I have seen statements along the lines of “Everybody should have kids except some very exceptional people who can influence the probability of AI catastrophe, who should invest their time into doing that”. Given the framework above, it seems that there is no reason for such a discontinuity of costs and benefits. Consider someone who believes the probability for an extinction catastrophe is low enough (so being born is still worthwhile), who believes that s/he can influence catastrophe probabilities a bit (and in principle would like to do so) and worries moderately about their sleep and health. If that person thinks that s/he would not enjoy having children then s/he should probably not do so, but if s/he is very enthusiastic about having children then should probably do so. It may seem strange that personal preferences should make a difference at all given that global catastrophe enter the decision. But if the person is indifferent given all other points including the global catastrophe, then personal enthusiasm will make the relevant difference.
The demand for a bright halo
People may confuse the three different perspectives when discussing the overall question. For example, someone may answer to the question as follows: “Of course you should have children! Life is good, and if you don’t have kids, then humanity will fade away.” Yes, this may be the case, but it is not an answer to the “child’s perspective” question or the overall question. However, this confusion of the perspectives may possibly not just be intellectual negligence.
Wikipedia says: “The halo effect (sometimes called the halo error) is the proclivity for positive impressions of a person, company, country, brand, or product in one area to positively influence one’s opinion or feelings. The halo effect is “the name given to the phenomenon whereby evaluators tend to be influenced by their previous judgments of performance or personality.” The halo effect is a cognitive bias which can prevent someone from forming an image of a person, a product or a brand based on the sum of all objective circumstances at hand.”
It seems plausible that there is a demand for something like a halo effect—a demand for a motivated belief—for decisions with ethical consequences.[5] Suppose you make a decision that affects other people. Then you may want to believe that the decision is not only a good decision in the sense that benefits to you and the other person outweigh the costs to you and the other person, but you may want to believe that there is no doubt that the decision is good even if you only consider the costs and benefits for the other person. Otherwise you might later have to tell the other person: “Well, I knew the decision would not be positive for you, but it was so good for me that it outweighed your costs”. That, however, does not sound like a good thing to explain to anyone.[6] So you may want to avoid it by lying to yourself about if you have decided to have children purely based on or even based on total (but with ).
Moreover, this failure mode is also valid for other conflations of arguments. You can say “having kids actually makes me more productive because it forces me to work more efficiently”. That’s fine. Maybe it is even true. But would you be open to it not being true? Or was your decision fixed before you even thought of that argument?
Alternatives
Depending on the relative strength of different of the above-mentioned factors, you may also consider alternatives to having children, like the following alternatives that are mentioned in LessWrong comments:
Mentoring if you want to simulate parenting or spread knowledge or memes.
Running clubs for kids.
Babysitting for other people.
Living in group houses and taking care of other peoples’ children there.
Some of these may also be good as experiments to check whether you would enjoy having children. Being foster parents or adopting a child could also be mentioned. The alternative of just taking your time and money to save other children or people in general and being proud of that is also relevant. Another alternative that is mentioned in LessWrong comments is:
Sperm donations if you want to spread your genes, or help other people have children.[7]
Conclusions
With this post, I hope to give you an overview about the SYHC decision influences, and what considerations it can be based on. The “net benefit” approach will not work in the way of really summing up effects, but it may be a useful framework nonetheless. I also listed some ways how rational decisionmaking of this kind of decision can fail.
What I do not include is an answer on the object level, that is, should you have children? One reason is that everything depends on preferences and individual life situations. Another reason is that I find many questions here hard to answer, and from my reading of family-planning posts not only on LW but also some other websites/forums, I am surprised how simple some people seem to find them.
I included some direct links to sources, and maybe I will add more later on. However, you can usually find them in the posts and discussions linked in the survey.
The impact of climate change on future life quality seems to keep people from having kids, but climate change is not regarded as such a relevant topic in the rationality community or the EA community, compared to the GCR topics I listed. wSee Please Don’t Give Up On Having Kids Because Of Climate Change.
Some people in the communities worry about low fertility rates. To argue that people should have more kids seems to be much easier if you can genuinely say: “Being born is great!” And yet, a clear and convincing message like this does rarely seem to be part of the discussions.
Should you have children? A decision framework for a crucial life choice that affects yourself, your child and the world
The need for a structured framework for deciding whether you should have children
In the rationality community and in the EA community, it is normal to analyze all kinds of decisions in detail, and this of course includes their ethical consequences—that is, the impact of decisions on other people. Yet even though there are a number of forum posts on the decision whether to have children, they often focus on the effects on personal productivity and resources or personal happiness of the potential parents. There are also some ethical considerations, but the effect of being brought into this world on the child itself receives surprisingly little attention. In general, there does not seem to be a place where all the factors of the “should you have children” (SYHC) question are brought together in structured way. This can lead to implicit normative assumptions and vagues ideas heavily influencing discussions.
Deciding whether to have children is one of the most important decisions in your life as it will require a large amount of your time and financial resources and will likely be a life-changing experience that will transform your worldview and preferences.[1] Of course, not having children can also drain your resources because, depending on your personality and values, you might feel incomplete and unhappy.
With this post, I aim to organize the SYHC discussions and offer a framework for the considerations that are necessary for them, in order to enable a more structured and rational discussion for those who are interested in the topic.
First, I discuss how to approach the question as a rational decision, and discuss the considerations that should, arguably, be part of a rational decision on whether to have kids. I split this into considerations regarding the parents, the child, and the world, discuss how these considerations relate and which role time horizons, discounting and timing play. After an interim summary, I discuss failure modes of rational discourse and rational decisionmaking in this context, with one particular fallacy that I suspect plays an important role in the SYHC decision. I also mention some alternatives to having children.
My conclusions do not include an object-level answer to the question whether you should have kids. In particular, this post is not intended to say that you should not have kids or that you can easily sum costs and benefits for this decision. I’d welcome discussions, including about empirical insights into the costs and benefits, in the comments to this post.
The decision of whether to have children as a rational decision
Suppose you contemplate one of the most important decisions of your life, namely, whether to have children. Like any other important decision, it will have positive and negative effects—costs and benefits -, and it comes along with uncertainty, possibly including uncertainty about your own preferences. Considering all these trade-offs can be challenging, and it can be even more challenging to reduce the uncertainties given that many people have very strong opinions about the topic (and may understand your questions as an attack on their decisions or values), nearly everybody has implicit (ethical or empirical) assumptions, and many other people may not find the aspects even discussable that you care about. However, as with other decisions, if you are really unsure about how to decide, it can be helpful to actually think about the factors that influence your decision. If you are not unsure, thinking about these decision factors can help other people who are unsure.
The simplest decision-making framework is:
”Do something if the expected net-benefit of doing it are positive, where net-benefit is benefit minus costs.”
I’ll ignore the uncertainty for the moment and just write for the decision criterion.
For many people, this sounds fine for business investments, but awfully cold and calculating in the case of children. However, if you consider “benefits” as positive deviations from the situation you would be in if you had no children and “costs” as negative deviations from that situation, it is a useful framework. Of course it can include the effects of having children on the child and on rest of the world.
where and are weights that depend on your preferences—that is, shows how much the child’s expected net-benefit influences the decision.
As a rational decision framework, this has limits, but it makes sense to first see how far we can get with the framework.
seems strange—after all, isn’t all of the of the parent? Yes, but there are more direct costs and benefits to the parents. This includes, for example, pregnancy sickness.
Note that what counts as costs and benefits may be very different from person to person, depending on your preferences, your biology or your opportunity costs. Therefore I will not list costs and benefits strictly separately, because many factors can be costs or benefits depending on your preferences. (For example, spending time watching cartoons with children can be fun for some people and annoying for others.)
For many of the factors discussed below, your personal preferences will matter a lot, and obviously your expectations about your personal situation and your life’s development are important. However, with respect to your (potential) child’s life and your child’s effects on the world, global developments are very important as well.
Social pressure, approval or disapproval can be part of the . This is also an example that you can influence some costs or benefits by changing your environment or having different (maybe more stoic?) attitude about them. At the same time, it is an example that there may be costs that you can only shape but not avoid—if you feel social pressure that you should have a child, then keep in mind that you will also feel social pressure with respect to your parenting decisions once you have a child.
The framework does not imply or require that you make this decision alone. Maybe you and your partner are thinking about these issues, maybe you have completely different opinions, for example about priorities, preferences, expectations.
It is important to remember that all of the factors influencing your decision are uncertain. You can invest time in personal research to become more certain about them, but the best you can get is subjective probabilities.
With all this in mind, let us consider the different categories in more detail. The lists are open for further additions, though they also try to include everything that has ever been discussed in this forum; see the survey “Should you have children? All LessWrong posts about the topic”.[2]
The costs and benefits for the parents
These are points to consider from the parents’ perspective:
How much do you like children? Do you like “the feeling of a soft warm baby in your arms”, spending time with children, playing with children? How much would you enjoy being parents? How much joy do you expect to get from playing with the kids, experiencing their growth, watching them be kids and grow up? How much would the less pleasant tasks of parenting burden you?
How much do you expect the same to be true for becoming grandparents when (and if) your children have children?
Do you expect that having children will be an interesting experience and satisfy your curiosity? This may include being curious how being a parent changes yourself, including the massive impact of hormonal changes on your mind—however, others may think this expected change will destroy part of your identity
How do you value the time you will (have to) spend with your children compared to other ways to spend your time? Children are very time-intensive and limit your autonomy with respect to your own time. They are a long-run obligation and thus a strong commitment.
Note that this does not constitute a cost with respect to the time that you actually want to spend time with your children, but it makes sense to think of this as “benefit = time spent with my children”, “cost = forgone trek crossing the Alps”. Would you travel the world if you had no kids? Would you spend your days partying? Would you save people from starvation? Would you solve AI alignment? This is a more important point for people who have very strong wishes how they would like to spend their time, or feel obliged to spend their time in certain ways.
However, note that time usage is not always rational. Instead of preparing packed lunches for your children, you might just binge-watch a series on TV and regret it later.
The amount of time required may depend on whether you have support, such as your own parents living in the same village, or if there are good kindergartens.
You may also believe that children are even more time-intensive or costly than it would reasonable, due to social pressure or wrong assumptions about your effects on the child’s development.
Raising children is financially costly.
What would you spend your money for if you had no children? The cosniderations are similar to those with respect to time.
This argument is sometimes made in particular noting that you could spend the money for altruistic causes. However, it also applies to people who just want to spend the money for themselves. Consider however you would spend the money if you had no children.
You much do you expect exhaustion / loss of sleep to bother you or to affect your health? (I don’t want to overstate this—people are surprisingly adaptable, and you can optimize this, like many other things. But it’s worth doing your own research here.)
How much do you expect having children will affect your health? (Children spread pathogens. But maybe you are more active because you have children, or your children may care for you when you are old which is good for health then.)
Monetary and time costs are not completely separable because you can, to some extent, convert one into the other. The same is true for e.g. time and health/sleep or money and health/sleep.
Do you expect having children to increase or decrease your productivity?
Maybe you will fulfill a fundamental desire that would otherwise haunt you and keep you from working productively.
Maybe the time scarcity makes you prioritize better.
Maybe your motivation (to create a good world for your children or earn money for them or whatever) increases.
Or maybe you are more distracted and no longer able to focus.
Do you expect having children to increase or decrease your short-run happiness / life satisfaction for some other (maybe unclear) reasons?
Do you expect having children to increase or decrease your long-run happiness / life satisfaction for some reasons (e.g., expecting pride or companionship)?
How much do you expect to enjoy “creating” a new life, or expect that you enjoy nurturing, educating, parenting their children? How important is it to you that there will be beings that are relatively similar to themselves, or whose personality you at least strongly influence? Do you expect to feel emotional satisfaction or a feeling of meaning from creating a life, seeing the children grow, being responsible for their success? Do you expect that this gives you the feeling of doing something “real” that you otherwise would not have?
Does your life partner does or does not want to have children, and you want to do something that he/she wants?
Do you think that by having a child you can avoid societal or peer pressure? Or do you expect additional disapproval if you have a child, or strong prescriptions by the government, society or your peers on how to raise your children?
Do you think that having children gets others to take you more seriously?
Do you expect to feel insecure/anxious? Do you expect to feel unable to raise kids? (If so, you may want to think more generally about how to reduce your anxiety. Or it might be helpful to know that you can get advice on raising children and that you will grow with the task.)
Do you fear status competitions that come along with stress and even more costs including more time costs? Note that all time and cost considerations, but also your stress level, is influenced by social expectations, in particular if you do not consciously decide how much to care about them.
Many benefits and costs (in particular, time cost, monetary cost, cost in terms of energy, health, but also feeling insecure or anxious) may be influenceable to some extent by planning, optimizing, coordinating with like-minded people, just talking to people, timing (e.g. having children at a time of your life when your own parents may be able to help because they retired but are still healthy).
The costs and benefits for the child
One topic that I don’t think is discussed enough is the impact of being born and brought into life on the child itself—the child’s perspective. Do you think that children born today can expect a good life today and in the future if they are raised by people like you?
The general quality of life depends on many things like the child’s family (including yourself), your income, the country you live in, the child’s biology and health etc.
Arguably, most people like to be alive. That most people don’t commit suicide is sometimes mentioned as evidence (though it is not clear to me how strong this evidence really is). Nonetheless, it seems plausible to me that most people like to live. Moreover, it seems that with today’s technology and resources, life could be very beautiful for everyone if manking coordinated better. Life can be improved on both the micro and macro levels. If you are reading this text, then you are a possibly a kind of person that will be able to read texts on how to do this.
It is sometimes added that people are better off now than they were in former times when people also had children, even in times like the Thirty Years’ War. However, this does not add anything to the argument. People decide differently, maybe they just do what everybody does, maybe they believe God commands them to have children. That you exist does not mean that your ancestors had a good life or even that they made good decisions. This confuses the “is” and the “ought”. Similarly, it does not add much that people in miserable situations today also have kids. It is, at best, a statement about the parents’ decisions. (Maybe and alternatively, the statement is a way of saying that contributing to the future of humanity is really important, but then people should say that more clearly.)
Is it good to be born even if you expect the world to end in a catastrophe in a decade? The question of “How much do you expect that your children will have happy and fulfilled lives?” would be strongly influenced by a catastrophic-risk expectation. The question is linked to the question of general expectations of humanity’s future. Thus, discussing whether the world will be worthwhile for children to live in should help to clarify what to expect from the future for yourself and how to prepare for it. Strangely, the relationship between expectations about Global Catastrophic Risks (GCR) like nuclear war, pandemics, and unaligned artificial superhuman intelligence[3] and expectations about a child’s life quality are rarely discussed even in a community that assumes that GCR have relevantly high probabilities.[4] It seems to be very important what kind of catastrophe the parents expect. For instance, if they expect an “everybody suddenly gets switched off by Nanobots” scenario, they may think that a 5-year life is positive, but it is possible to imagine scenarios that include much more suffering.
There are people who assume that living is bad in general. (For a rejecting discussion of this position, see for example “Antiantinatalism” by Jacob Falkovich.)
There are also people who assume that the future will not be worthwhile to live in; I have not really seen very good and credible discussions of expectable future life quality (with or without transformative AGI).
However, it is clearly also true that no life is only and always pleasant. At the same time, there is no reason to assume that is a binary. If you know that Utopia will soon start (whatever you imagine Utopia to) , then life will be positive on average; if you expect a dystopian S-risk scenario that lasts forever, then life will be probably something to avoid.
The costs and benefits for the world
These are points that affect the world—the “world’s perspective”, maybe the humanity’s perspective or the perspective of sentient beings.
Do you think that the existence of (additional) people is good for the world? If you think that the existence of people is good then you may contribute to their future by having children.
There are people who think that having children is good for the world, even if the same people think that having children is bad for the parents.
Arguments why the existence of people is good for the world may include that younger people are necessary to care for old people, or that young people are good for the economy. Do you think that people have positive or negative externalities? Is there strong resource scarcity for humanity, or do additional children add so much to technological progress that their marginal contribution is positive?
Do you think it would be good for the world if people like you or you in particular should have children (selective pro-natalism)?
Do you think that your children would make the world a better place?
Do you assume that it is better if “wealthy, educated and responsible people have more kids”? (This argument holds for wealthy, educated and responsible people—however, you may like other characteristics of yourself that may support similar arguments).
Do you want to contribute to the survival and spread of certain kind of preferences/memes/culture? (There are people who also see children as an “investment in AI research”)
Are you rich and do you believe that rich people should have more children in order to increase equality?
Altruistic actions you give up because you have too little time or money are not part of because they are no direct effects of your child; instead they are part of , because the child changes your time or money usage.
It seems possible that the quasi-altruistic motivation of contributing to the future of humanity by having children may increase due to a catastrophic-event expectation, because maybe the number of people surviving the event is higher if there are more people before the event. But this depends on the kind of event. The “more people will be left” thought would definitely apply if some catastrophe is expected to kill a fixed absolute number of people or even a fixed share of the world population, but would not apply if for some reason the number of people expected to survive is fixed or it is clear that only people living in a specific unreachable area of the world will survive.
Relations between the three groups of costs and benefits
Many positive feelings for the parent () would probably vanish if they expected , so treating them as separable has limits. To some extent that may even be true if parents expected .
Even very direct costs may depend on the meaning attached to them. Spending your time taking your child to a guitar teacher and waiting until the lesson is over may be costly, but it would be much more costly if you did it for some stranger’s child, whereas you may not find it costly at all if you attach meaning to it, like contributing to your child’s ability to enjoy music in the future. Similarly, you may find the sleep loss less costly if you think that your child will have a happy future and also contributes to the benefit of the world.
Parental feelings that they have a “duty” of having children or that this is part of a “fulfilled life” may be part of but that only makes sense if .
Assuming that having children is good for the future of humanity () depends on the assumption that life is good in general (i.e. not for your child in particular, but for children on average, at least for sufficiently many generations in the future).
or depend on your effort and investments (but maybe not as much as you think), which influences .
Notes on time horizons, discounting and timing
Note that all considerations depend on your time horizon.
Expecting a long life means that you can expect a greater , because some of your benefits only accrue when you are older.
If you expect your child to live longer, then will likely be larger, given that it is positive at all. Though some argue that happiness is highest for people when they are children and decreasing over a person’s lifetime, additional years add value if the for these years remains positive.
If you expect humanity to continue to exist for longer and its existence to remain positive, then investing effort now and contributing to the future development of humanity is more valuable. Expecting a high probability of an existential catastrophe for humanity will not only reduce , but also and .
However, if you think that humanity is about to get destroyed by a catastrophe and you can do something against it, then this decreases (because it increases your opportunity cost).
If you strongly discount the future, all of this may matter less—but this is also true for anything positive that would happen in the future (including later in your own life). This would justify having children if you really just want the feeling of having children right now.
Notice that a five-year time horizon (with an end-of-the-world endpoint) does not necessarily increase your opportunity cost for these five years. Sure, if your life goal was to visit the Louvre and you expect the world to end in five years, then you need to visit the Louvre now and not when your children are 18. However, if your plan was to build a big business, that will seem much more difficult in five years than it will in 20 years (and what’s the point of the goal then anyway), so you may not stick with that goal anymore.
In addition, your opportunity cost over the five years depends on whether or not you believe you can influence the likelihood and magnitude of the disaster. Some people seem to believe that having children greatly increases their focus or motivation, which may offset the necessary time investment.
Relatedly, costs and benefits may depend on timing and the order in which you do things in your life.
In general, you need a partner and biologically you can only have children during a certain phase of your life. The probability to be able to have children decreases when you get older. You may be too optimistic when you are young.
If you have children earlier, you will have more time overlap with them, which seems good for you and for the children (and for your potential grandchildren).
If you have children later in life, you may be more financially secure and have more life experience.
Some people argue that one can do other things first because in the (sufficiently near) future a superintelligent AI will solve all problems by either making humans immortal (which would then extend all time horizons enormously) or at least by overcoming fertility problems (including retroactively restoring fertility). This argument can of course also be made in a weaker form: perhaps you want to wait instead of having children now because you hope for technological changes that will improve the lives of your children.
Summing up the costs and benefits
Considering the costs and benefits listed above (and adding costs and benefits that are not yet included in the lists), it is possible to think about the SYHC decision in a structured way. Literally adding and subtracting will not work, because you would have to attach some monetary or similar value to the costs and benefits and that will often not work or not feel right. However, you can use such lists as a help to understand the consequences of your decision.In theory, you could write down your expectations for your life with children and your life without children, then compare these options and see the differences in each area. Conversely, ignoring costs and benefits does not mean that they do not exist.
You can also use this framework to discuss your decision with other people. If your partner has a different opinion, but also thinks that discussions are a good approach to decisions, then this may be a good starting point.
Even if your decision is already set in stone because according to your preferences, one of the points above is more important than all other points together, then you can use this framework to understand its consequences in other areas or understand people you disagree with.
Failure Modes of Rational Discourse
It is possible that rational discourse fails. The issue at hand seems particularly vulnerable to a failure of rational discourse. Some reasons for this:
People can feel very strongly about the topic and therefore not be willing to engage in rational discourse at all.
People sometimes pathologize other people, for example they claim that their worries must exclusively come from emotional distress instead of discussing their worries.
Parenting decisions are usually decisions that are made with your partner, so this requires communication. This kind of communication may be difficult because so much depends on it.
Some say that “whether to have children isn’t each other’s business and pressure against doing normal human things like this is net socially harmful” which may also lead to misunderstandings or fallacies. So be aware that some people may perceive discussing these issues as intrusive.
Some people think that discussing reasons and arguments on the topic is nonsensical. For instance, someone may think: “At some point in your life, you just have kids. It’s just what you do. That’s the way life is!” This kind of explanation seems similar to the “fulfilled life” or “duty” arguments already mentioned, but it explicitly avoids considering the arguments. This attitude may seem incredibly rigid, but at the same time, it may be very comfortable. If this is how you approach decisions, you do not have to weigh costs and benefits, and have you one thing less to think about. Most importantly, you will never feel responsible for any consequences. So even though this way of deciding may not be rational, it may still be good for some people. However, these people may not have a way to find out whether this is the case, and they may very well change their mind later on and regret their decisions (this of course may also be the case for people who approached their decisions in a more rational way).
“If someone tells you not to have kids then that is crazy and you should not even consider arguments about it. The voluntary human extinction movement argues for not having kids and that is obviously crazy.” Well, maybe there are people who do not know and understand your position yet? If so, then you can do them a favor by explaining to them why telling people not to have kids is crazy.
To discuss the considerations that influence the having-kids decisions, I included examples like GCR and suffering risk. There are more mundane fears like “Will I put my kids into a world that completely lacks meaning and perspective?” which may seem even more crazy—or at least that is the impression of answers I saw in discussion of such questions. I still could not explain to myself why these worries are not justified. Even a “positive” AI future, as little as I can imagine it, seems worrying to me. However, I tell myself that this is just a typical human reaction—many people reject any kind of change.
Failure Modes of Rational Decisionmaking
An even more basic failure mode is that of rational decisionmaking. Why may that failure mode be relevant here?
Emotions (urges, desires, feeling insecure) can be very powerful. Hormones etc. can be very strong, in particular during some years of your life. This can give rise to rationalization.
Having children may completely change your preferences, so it can be a transformative experience. The decision may feel very different after you made it.
We do not completely know our own preferences. Therefore, it makes sense to consult statistics on happiness / life-satisfaction; but people question whether happiness is actually the same as long-run preference satisfaction.
Happiness research seems messed up. Are people nowadays really as much happier than 100 years ago as you would expect based on your knowledge about economic growth? Also, how do you really evaluate a life’s happiness given the “Two selves” that Kahneman explains in “Thinking, Fast and Slow?
Time-Inconsistency: Another way that decisions are irrational is time-inconsistency. Consider someone who describes his or her decision like this: At some point in your life, you feel overwhelmed by some vague but extremely strong desire to have kids. That is because you just want it, or because of your hormones, or whatever kind of perspective or level of explanation you prefer. This suddenly overrides all of your previous plans for your life. If preferences are like this, they are immune to rational considerations. This is similar to the desire of a drug addict to get more heroine; you can tell him that drugs are, all things considered, bad, so he should be happy that there are none available; but he may not be happy. Actually, there is a very specific way your preferences have to be over your lifetime that does not give rise to time inconsistency (adding up utility for every moment of your life, exponentially discounted or not discounted at all), and it is not clear that brains work that way. Making consistent decisions over time is hard and nobody prescribes what your preferences should be.
Regret and alternative realities: Which kind of regret disutility do you fear most? Do you fear that people say “I wish I had never been born!” (probably not, you can always say “I did my best and all other people had kids) or do you fear that on your deathbed you think “I wish I had had kids”? In the latter case, on your deathbed, you might compare murky reality to an optimal hypothetical life and therefore feel a lot of regret. This motive of avoiding regret may not be compatible with rational considerations because the object of regret is itself not grounded in reality (or in realistic expectations over what could have been).
Confirmation bias: People may stop taking in new information once they have made a decision, because this information may question their decision, which creates a bad self-image.
I have seen statements along the lines of “Everybody should have kids except some very exceptional people who can influence the probability of AI catastrophe, who should invest their time into doing that”. Given the framework above, it seems that there is no reason for such a discontinuity of costs and benefits. Consider someone who believes the probability for an extinction catastrophe is low enough (so being born is still worthwhile), who believes that s/he can influence catastrophe probabilities a bit (and in principle would like to do so) and worries moderately about their sleep and health. If that person thinks that s/he would not enjoy having children then s/he should probably not do so, but if s/he is very enthusiastic about having children then should probably do so. It may seem strange that personal preferences should make a difference at all given that global catastrophe enter the decision. But if the person is indifferent given all other points including the global catastrophe, then personal enthusiasm will make the relevant difference.
The demand for a bright halo
People may confuse the three different perspectives when discussing the overall question. For example, someone may answer to the question as follows: “Of course you should have children! Life is good, and if you don’t have kids, then humanity will fade away.” Yes, this may be the case, but it is not an answer to the “child’s perspective” question or the overall question. However, this confusion of the perspectives may possibly not just be intellectual negligence.
Wikipedia says: “The halo effect (sometimes called the halo error) is the proclivity for positive impressions of a person, company, country, brand, or product in one area to positively influence one’s opinion or feelings. The halo effect is “the name given to the phenomenon whereby evaluators tend to be influenced by their previous judgments of performance or personality.” The halo effect is a cognitive bias which can prevent someone from forming an image of a person, a product or a brand based on the sum of all objective circumstances at hand.”
It seems plausible that there is a demand for something like a halo effect—a demand for a motivated belief—for decisions with ethical consequences.[5] Suppose you make a decision that affects other people. Then you may want to believe that the decision is not only a good decision in the sense that benefits to you and the other person outweigh the costs to you and the other person, but you may want to believe that there is no doubt that the decision is good even if you only consider the costs and benefits for the other person. Otherwise you might later have to tell the other person: “Well, I knew the decision would not be positive for you, but it was so good for me that it outweighed your costs”. That, however, does not sound like a good thing to explain to anyone.[6] So you may want to avoid it by lying to yourself about if you have decided to have children purely based on or even based on total (but with ).
Moreover, this failure mode is also valid for other conflations of arguments. You can say “having kids actually makes me more productive because it forces me to work more efficiently”. That’s fine. Maybe it is even true. But would you be open to it not being true? Or was your decision fixed before you even thought of that argument?
Alternatives
Depending on the relative strength of different of the above-mentioned factors, you may also consider alternatives to having children, like the following alternatives that are mentioned in LessWrong comments:
Mentoring if you want to simulate parenting or spread knowledge or memes.
Running clubs for kids.
Babysitting for other people.
Living in group houses and taking care of other peoples’ children there.
Some of these may also be good as experiments to check whether you would enjoy having children. Being foster parents or adopting a child could also be mentioned. The alternative of just taking your time and money to save other children or people in general and being proud of that is also relevant. Another alternative that is mentioned in LessWrong comments is:
Sperm donations if you want to spread your genes, or help other people have children.[7]
Conclusions
With this post, I hope to give you an overview about the SYHC decision influences, and what considerations it can be based on. The “net benefit” approach will not work in the way of really summing up effects, but it may be a useful framework nonetheless. I also listed some ways how rational decisionmaking of this kind of decision can fail.
What I do not include is an answer on the object level, that is, should you have children? One reason is that everything depends on preferences and individual life situations. Another reason is that I find many questions here hard to answer, and from my reading of family-planning posts not only on LW but also some other websites/forums, I am surprised how simple some people seem to find them.
For the preview picture, I used Children’s Games bei Bruegel.
It is even possible that something about your children makes you become a political activist and put billions of dollars into that acitivism.
I included some direct links to sources, and maybe I will add more later on. However, you can usually find them in the posts and discussions linked in the survey.
The impact of climate change on future life quality seems to keep people from having kids, but climate change is not regarded as such a relevant topic in the rationality community or the EA community, compared to the GCR topics I listed. wSee Please Don’t Give Up On Having Kids Because Of Climate Change.
Some people in the communities worry about low fertility rates. To argue that people should have more kids seems to be much easier if you can genuinely say: “Being born is great!” And yet, a clear and convincing message like this does rarely seem to be part of the discussions.
Is there a reversed version of “The Fox and the Grapes”?
For the same reason, most people may not be inclined to weigh costs to animals and benefits to themselves of eating meat.
I do not know whether there is sperm-donation scarcity.