What this covers

Bryan Caplan of George Mason University joins Russ Roberts to discuss his book Selfish Reasons to Have More Kids, advancing a provocative claim grounded in behavioral genetics: that adoption and twin studies reveal parents have surprisingly small lasting effects on their children's long-run outcomes. The argument spans health, intelligence, education, income, character, and values. If this research holds, Caplan contends, much of the anxiety and labor parents invest to shape their children's futures is unnecessary, and the central implication follows—parents can relax their efforts, focus on activities both they and their children enjoy, and rationally choose to have more children because the effective cost of raising successful kids is lower than convention suggests.

The conversation moves through the empirical ground: specific adoption studies showing that wealth of adoptive parents barely predicts children's eventual income, twin research on lifespan and smoking patterns, and critiques of conventional child-development research for ignoring heredity. Caplan argues that much parental influence amounts to a high-variance gamble, as likely to backfire as to succeed, and that the one robust effect parents do have is on the relationship itself—whether children feel their parents were kind and present. He challenges assumptions about childhood safety (stranger abduction remains vanishingly rare), discipline (effective only while maintained), and the quality-quantity trade-off in family size. Roberts raises the concern about population's environmental cost, which Caplan addresses by arguing that additional people generate net positive externalities through ideas and market expansion. Throughout, Caplan separates what genuinely matters long-term from what parents do to satisfy status concerns or satisfy themselves, and recommends parents aim for "decent roommates today" rather than permanent transformation.

Sharpest takeaway

Caplan argues that adoption and twin research shows parents have surprisingly little long-run effect on how their kids turn out, so much of the unpleasant 'investment' parenting is unnecessary—meaning kids are cheaper and more enjoyable than people think, and parents should have more of them.

  • Adoption and twin studies separate nature from nurture and consistently find small or zero long-run parental effects on health, intelligence, education, income, character, and values
  • If pushing kids hard doesn't change long-run outcomes, parents can drop unpleasant parenting and focus on enjoyable activities and a good relationship
  • Lower effective cost of raising successful kids implies people should rationally choose to have more children

The claims · ranked31 claims · weighted by value

This asset isn't compiled yet

You're seeing its claims, ranked. Compile it to build the argument threads, weight them, and check each claim against your library — the full view.

0.81

Stranger abduction is one of the most over-feared risks: it is roughly a one-in-a-million-per-year event, and a child is about a hundred times more likely to die in a car crash—so refusing to let a responsible child walk to the mailbox or stay home alone while inflating abduction fears is statistically incoherent.

factualhigh valueestablishednovelty 3/4durability 3/4· Bryan Caplan

“stranger danger... it literally is a one in a million per year event”

0.79

The same adoption/twin data showing small parental effects is equally consistent with a theory in which parents are about equally likely to change their kids in the way they want or to have it backfire and produce the opposite result (e.g., a religious parent producing an atheist), making parental influence a high-variance gamble rather than reliably zero.

causalhigh valuecontestednovelty 4/4durability 3/4· Bryan Caplan

“the data is equally consistent with a theory where parents are about equally likely to change their kids in the way they want or have it completely backfire and get the opposite of the result they desire”

0.78

Much child-development research that does not use adoption/twin methods is unreliable not necessarily because of fabrication but because it consistently ignores heredity—e.g., concluding daycare helps or harms kids while failing to consider that parents who use daycare may differ genetically, so the children would have differed anyway.

normativehigh valuecontestednovelty 3/4durability 4/4· Bryan Caplan

“my main complaint is... they consistently ignore something that we have strong reason to believe is important which is reddit e”

0.76

Despite nostalgia for the 1950s as idyllic, the death rate for children then was about five times higher than today, driven mainly by disease and secondarily by accidents; deaths from crime and suicide are slightly up but remain microscopic, so the modern world is comparatively much safer for kids.

factualhigh valueestablishednovelty 3/4durability 2/4· Bryan Caplan

“even though we think about the 50s as being an idyllic age the death rate for kids was about five times as high as it is today”

0.74

In the Sacerdote adoption study, being born into a richer adoptive family did not increase the child's educational attainment, and children adopted by the very richest families grew up to have the same average income as those adopted by the very poorest families—implying growing up in a rich household does not transmit the secrets of success or pay off via nepotism by adulthood.

factualhigh valuecontestednovelty 4/4durability 2/4· Bryan Caplan

“adjusting for everything else being born into a richer family did not help how much education you got it was not the case that richer families bought educational success for their kids”

0.73

Gary Becker's quality-quantity trade-off—the assumption that having more kids means worse (lower-investment) kids—was assumed without much evidence, and adoption/twin research even around 1980 already suggested the trade-off is much weaker than commonly thought.

factualhigh valuecontestednovelty 3/4durability 3/4· Bryan Caplan

“what I disagree with rather is Becker's assumption that there is a strong quality quantity trade off he basically just took it for granted”

0.73

The one large, lasting effect parents do have is on how their children feel about and remember them; a Swedish twin study found that, regardless of genetic relatedness, people who grew up in the same household had similar views of whether their parents were kind, spent time with them, and treated them well—so parents who push kids hard on dimensions that don't matter long-run risk damaging the relationship that does.

causalhigh valuecontestednovelty 3/4durability 3/4· Bryan Caplan

“there's well one very meaningful fact the parents have on their kids and this is how their kids feel about and remember them”

0.73

On the 'Ask the Children' survey, children's main complaint about their parents was not that parents don't spend enough time with them (a rare complaint) but that parents are too tired, too stressed, and have bad tempers—so parents would earn better grades from their kids, and have an easier time, by cutting themselves slack and relaxing rather than dragging themselves to extra activities while exhausted.

factualhigh valuecontestednovelty 3/4durability 3/4· Bryan Caplan

“most parents assume that their children child's main complaint is my parents don't spend enough time with me that was a very rare complaint... the main complaints the kids have... is their parents are too tired too stressed and... have very bad... bad tempers”

0.73

Trying to fundamentally change another person—whether a spouse one marries to change, or a child one treats as the exception to the rule that you can't change people—is naive and likely to backfire and breed resentment; the realistic aim is to make your children into decent roommates today and to be the person who is kind and good to them.

normativehigh valuecontestednovelty 3/4durability 3/4· Bryan Caplan

“the idea of marrying another person in order to change them... is one of the absolute most foolish ideas that you can adopt and... thinking that your child is very different from all... humans this is the one human that it does make sense to try to totally remove them that is... equally naive and equally likely to blow up in your face”

0.73

The SAT is effectively an IQ test (it correlates with named IQ tests as highly as IQ tests correlate with each other); while practice can boost scores on any IQ test, that effect generally fades out after prep, and the practice gain is specific to the particular test rather than transferring to a novel test.

factualhigh valuecontestednovelty 3/4durability 3/4· Bryan Caplan

“contrary to what the SAT said officially says where the ETS says it is an IQ test it correlates as highly with IQ tests that are called IQ tests as IQ test coral.you each other that is what the SAT is”

0.73

Once you examine adoption and twin evidence, the surprising punchline is that parents have very little effect on their children's long-run outcomes across nearly everything parents care about: health, intelligence, happiness, educational success, occupational/financial success, character (honesty, kindness, work ethic, discipline), and values (religion, politics, family values).

causalhigh valuecontestednovelty 3/4durability 3/4· Bryan Caplan

“once you take a look at the adoption twin evidence the big punch line is very surprising is that parents turn out to have surprisingly little effect on long-run outcomes on how their kids turn out”

0.73

Part of why people push their kids and have fewer of them is the desire to use children's achievements as a kind of status accoutrement—to fit comfortably into social circles where kids attend certain schools and achieve certain things—which is more about boosting parents' current status than about an investment in the child's future.

causalhigh valuecontestednovelty 3/4durability 3/4· Russ Roberts

“the pleasure they get in wearing their children as ornaments this is a really appalling idea I think but I think there's some truth to it”

0.73

If parents have no large long-run effect on their kids, then much of the unpleasantness of current parenting is unnecessary, so parents can in good faith stop pushing their kids so hard and focus on enjoyable activities—and because the kids you want are therefore cheaper than you think, you should have more of them.

normativehigh valuecontestednovelty 3/4durability 3/4· Bryan Caplan

“if it really is true the parents are not having a large long-run effect on their kids then a lot of the unpleasantness that we're currently parents currently experience actually is really not necessary you can responsibly and in total any good faith as a parent I'll stop pushing your kids so hard and focus more on doing things that you enjoy”

0.73

In Bruce Sacerdote's study of Korean War orphans adopted by American families, a mother having one more year of education raised her adopted child's eventual schooling by only about five extra weeks—so a mom would need over 10 extra years of education (roughly two graduate degrees) to boost the child's attainment by a full year.

factualhigh valuecontestednovelty 3/4durability 3/4· Bryan Caplan

“if a kid got adopted by a family where the mom had one more year of education that kid on average finished about five extra weeks of education”

0.73

The highest-quality twin studies (using Scandinavian records going back to the 19th century) find that parents have essentially no effect on how long their children live; the correct interpretation is not that smoking and diet don't matter but that parents have little or no effect on whether their kids smoke, with smoking running in families because of a genetic propensity passed from parent to child.

causalhigh valuecontestednovelty 3/4durability 3/4· Bryan Caplan

“the result that you get from the very highest quality studies is that parents actually have no effect on how long people live”

0.73

Adoption/twin researchers in psychology were disproportionately people who originally expected and wanted upbringing to matter and who feared social and professional sanctions (including tenure risk) for publishing heritability findings, yet published them anyway—which makes the surprising small-parental-effect results more credible rather than the product of motivated bias.

factualhigh valuecontestednovelty 3/4durability 3/4· Bryan Caplan

“a common profile of an adoption or twin researcher in psychology... is someone who didn't want the stuff to be true and felt like their other colleagues would dislike them and look down on them and possibly not give them tenure... and they publish it anyway”

0.73

Even a person who is merely an ordinary consumer (not an inventor) helps generate new ideas, because the size of the market determines the payoff to innovation—illustrated by the Gilligan's Island professor whose invention is worth $7 on a seven-person island but $7 billion if it can reach seven billion people.

causalhigh valuecontestednovelty 3/4durability 3/4· Bryan Caplan

“just being a regular consumer also tell also encourages the production of ideas”

0.73

Adoption/twin research implies a 'vaguely normal range' of parenting styles that are all about equally fine for long-run outcomes, so parents shouldn't beat themselves up over choosing one approach over another and should focus on activities that are fun for the whole family rather than ones that make either parent or child miserable.

normativehigh valuecontestednovelty 3/4durability 3/4· Bryan Caplan

“the lesson we really get from adoption twin research is just that there's a very wide vaguely normal range we're all parent parenting styles are about equally okay”

0.73

Adding one more person to the planet has a positive net externality rather than a negative one, because the main driver of rising living standards over the past 200 years is ideas, and ideas come from people; deleting half the names from music, science, technology, or business would have slowed human progress.

causalhigh valuecontestednovelty 3/4durability 3/4· Bryan Caplan

“the net externality of adding another child to the world is positive rather than negative”

0.73

Clear, consistent, mild discipline (e.g., the naughty corner) is experimentally shown to work and substantially improve even difficult kids' behavior, but only while it is maintained—if you stop disciplining, behavior reverts—so discipline should be aimed at making children decent roommates today, not at permanently changing who they will be at 30.

causalhigh valuecontestednovelty 3/4durability 3/4· Bryan Caplan

“clear consistent mild discipline really does work and even bad kids do respond to it and it does substantially improve their behavior”

0.73

Children can handle far more independence than we currently give them and don't need the level of supervision typical of modern parenting, so it is fine to let kids do their own thing for much of the day.

normativehigh valuecontestednovelty 3/4durability 3/4· Bryan Caplan

“the basic point that kids absolutely can handle more independence than we currently give... them is right they don't need the amount of superficially give them it's okay to let your kids just do their own thing”

0.73

Adoption studies disentangle nature from nurture because adopted children share no genes with their adoptive parents; if adopted kids grow up to resemble their adoptive families, that resemblance must be due to upbringing rather than heredity, and studying thousands of adoptive families reveals whether different family environments produce systematically different outcomes.

definitionhigh valueestablishednovelty 2/4durability 4/4· Bryan Caplan

“there's a special kind of family families that adopt in those families if the child grows up to resemble the family to adopts him you can't say heredity”

0.73

Twin studies measure heritability by comparing how much more similar identical twins (sharing 100% of genes) are than fraternal twins (sharing 50%); adding 50 percentage points of genetic similarity lets you estimate genetic effect, and crucially, also reveals how much room remains for parents to increase similarity further—for example, an identical-twin correlation of 0.85 versus fraternal 0.80 implies 50 extra points of genes only raises correlation by 0.05, leaving little explained by nurture.

definitionhigh valueestablishednovelty 2/4durability 4/4· Bryan Caplan

“there are two different kinds of twins they're identical twins... who share 100% of their genes... and then there are fraternal twins these are twins who only share half their genes”

0.69

Accident deaths fell most for younger children—whose safety parents can control—and fell less for older teenagers who decide for themselves, which is at least suggestive evidence that parental bossing on safety (seatbelts, etc.) genuinely reduces child accident rates.

causalhigh valuecontestednovelty 3/4durability 2/4· Bryan Caplan

“deaths from accidents fell fell fell the most for younger kids and then basically for older teenagers... those acts and rates fell... quite a bit less... yeah well those are the ages where the kids actually start deciding for themselves”

0.68

The Ferber 'cry it out' method—once a child is about three months old, letting him cry for five to ten minutes, then comforting and leaving, repeating until he sleeps through the night—is an experimentally tested technique that works for a large fraction of kids within a short time, potentially saving roughly two years and nine months of sleep deprivation per child, with no evidence it produces insecure or unhappy adults.

causalhigh valuecontestednovelty 2/4durability 3/4· Bryan Caplan

“there is an experimentally tested method of getting kids to sleep that really does work when tried it's called the film method”

0.66

Newer 'children of twins' research, which combines twin methodology with the fact that twins' children grow up in different environments depending on whom the twin married, tends to find somewhat larger effects of upbringing than other methods—though it remains methodologically contested (e.g., the spouse's genes may explain the result).

factualhigh valuecontestednovelty 3/4durability 3/4· Bryan Caplan

“there is a new cutting edge body of research on the children of twins... those studies generally do find a bit larger effect for upbringing than the other methods”

0.63

Parents today spend more time taking care of their kids than they did during the baby boom, largely because they believe they must do many unpleasant things to protect their kids' future success in a competitive world.

factualhigh valuecontestednovelty 2/4durability 2/4· Bryan Caplan

“it is true that parents push themselves very hard today actually they spend more time taking care of their kids than they did during the baby boom surprisingly enough but a lot of what they're doing is based on the idea that they have to do a lot of unpleasant things their kids in order to protect their kids future”

0.63

We are not running out of resources: over the past 150 years the inflation-adjusted price of food, fuel, and minerals has fallen by about 1% per year on average (despite periodic spikes), and air and water quality in the first world have improved for decades even as population rose—so genuine environmental problems should be handled with targeted measures like pollution taxes or geoengineering rather than by reducing the number of people.

factualhigh valuecontestednovelty 2/4durability 2/4· Bryan Caplan

“over the last hundred and fifty years the product the price of food fuel and mineral has... actually fallen by about 1% adjusting for inflation on average”

0.43

Within a culture at a point in time, people with more income tend to have smaller families, and richer societies have smaller families—an empirical regularity Becker explained via the quality-quantity trade-off.

factualestablishednovelty 1/4durability 3/4· Bryan Caplan

“it's certainly true empirically that richer societies have smaller families and people as they tend to get richer... within a culture people who have more income to have smaller families correct oh yeah yes that is true”

0.35

In Western countries roughly one-third of twins are identical and two-thirds are fraternal, and about 3% of the population are twins.

factualestablishednovelty 1/4durability 3/4· Bryan Caplan

“in Western countries it's basically one-third identical to third fraternal”

0.12

A child misbehaving and being punished shows behavior temporarily improves, but the key question is how long that improvement lasts.

factual· Bryan Caplan

“child's misbehaving you punish him his behavior temporarily improves but the question is how long does that improvement last”