The paper behind this story
The Price of the Marginal Child
Request the full whitepaper
Tell me who you are and where you work. I’ll send the complete paper.
Sent to the author directly. No list, no forwarding, no third parties.
The cost of a child is usually drawn as a ramp, a steady climb where each child costs about what the last one did. So we costed out the next child, one at a time, for three ways of living an Orthodox life: the house that child forces, the tuition it adds, the car it fills. The ramp turns out to be a staircase, the steps are not evenly spaced, and the tallest one sits in a different place in every community.
Why would one child cost more than the child before it?
In 2024 two economists published a finding with a startling title. Car seats, they argued, work as contraception. Not the seat itself, which costs a couple hundred dollars, but the fact that three car seats will not fit across an ordinary back seat. A third child forces a family to buy a bigger car, and that lumpy, once-off expense lands squarely on the decision to have a third child.
Nickerson and Solomon estimate the effect at roughly 8,000 fewer births a year, around 145,000 since 1980. The lesson is not really about cars. It is about the shape of cost. A family standing at the third child is not weighing an average. It is weighing a step, and a tall enough step can stop the climb.
If a family has already climbed the car-seat step, what step comes next?
The car-seat step that stops the average American family at three children is a step Orthodox families climb as a matter of course. The minivan is bought early and without drama. So the interesting question is not whether they clear the third step. They do. The question is what the next steps look like, at the fourth child and the fifth, and whether one of those is tall enough to end a family that wanted to keep going.
To answer it we build three staircases, one for each way of living an Orthodox life. A yeshivish world of large planned families and modest incomes. A modern Orthodox world of dual professionals and expensive schools. An out-of-town world of cheaper houses and school vouchers, where most established communities run their own schools and only the smallest have no high school within reach. Same question in each: how tall is the fourth step, and the fifth?
How tall is each step, and what is it made of?
Each block in the staircase below is one child, the third through the sixth, read left to right. The height of a block is what that child adds to the family's fixed budget every year, so a taller block is a more expensive child. Every block is stacked out of what the money goes to. The blue section at the bottom is housing, the bigger home that child forces. The orange section above it is tuition. The grey section on top is the car. The dollar figure printed on each block is that child's annual cost, the small family standing on the ground at the left is where the climb begins, and the four buttons redraw the whole staircase for a different community.
Start where the buttons open, on yeshivish, and watch the blocks get shorter as you climb.
So where is the tallest step in a modern Orthodox family?
Switch the staircase to modern Orthodox and one riser towers over the rest. The fourth child costs about $38,800 a year at the margin. Roughly $17,000 of that is not tuition at all. It is the house. Three children could share bedrooms in a way that four, split by the plain fact that boys and girls need separate rooms, cannot. The fourth child is the one that forces the move from a three-bedroom home to a four-bedroom home, and in Teaneck that move is expensive before a single tuition bill is added.
One additional child, eleven percent of a $350,000 income, arriving not as a slope but as a wall.
And then, tellingly, the fifth child is cheaper than the fourth, about $21,000, because the big house is already bought and the fifth child simply moves into it. The expensive decision in this community is not the fourth-versus-fifth that the paper set out to study. It is the third-versus-fourth. The wall comes early.
Does every community's staircase climb?
In the yeshivish staircase the steps get shorter as you climb. The third child costs about $12,600 at the margin, the fourth about $8,100, the fifth about $7,100. The climb eases.
Two institutions do this work. Bedrooms are shared more, three and four children to a room, so the house does not force an upgrade anywhere in this range. And the schools discount for siblings, sometimes capping what any one family pays, so each successive tuition is smaller than the one before it. The community has, in effect, built handrails into its own staircase.
If the same child costs so differently in each place, who is setting the price?
A child costs what a community's institutions make it cost. Three cards below, one staircase each, and the same three children in every card: third, fourth, fifth, left to right. The colours carry over from the big staircase, blue for housing, orange for tuition, grey for the car. All three cards are drawn to one shared scale, so a block twice as tall as another is twice as expensive, whichever card it is on.
Look only at the middle block on each card, the fourth child, and compare the three heights.
Three shapes, three answers. In one community the climb gets steeper as you go, in one it gets gentler, and in one the tallest step is hidden further down the road. Each shape is drawn by the schools and the bedrooms of the place, not by the biology of a fourth or fifth child.
Does the car we started with still matter at the fourth child?
Return to where we started. The car-seat step is real, and it fires here too, at the third child, when a family moves from a sedan to a minivan. The next car step, up to a twelve or fifteen-passenger van, does not arrive until the sixth child or so. Which means the fourth and fifth child sit in the gap between the two, in a minivan the family already owns.
The wall that stops the average family at three is behind them. The next one is not yet in front of them. On the road, the fourth and fifth child cost nothing at the margin.
This is a small point that matters. For a large Orthodox family the car is beside the point, and what decides the fourth and fifth child comes down to two costs, the house and the school, which is where this whole essay lives.
Can a step arrive years after the child who caused it?
Out-of-town communities are the cheapest place to have a fourth child. Vouchers cut the tuition, the houses are affordable, and a fourth child can cost under $4,000 a year at the margin. For most of them the bargain holds all the way through, because the established communities, places like Memphis, St. Louis, Columbus, and Cincinnati, run high schools of their own. The catch belongs to the smaller and further-out communities, the ones with a single school or none past eighth grade, where a family has no local option and a teenager has to be sent away to a dormitory yeshiva.
The chart below follows a single child from left to right, from the start of school to graduation, with the years marked along the bottom. The height of the shaded band is what that child costs the family out of pocket in that year, so a taller band is a more expensive year. The dark dot sits at ninth grade.
Follow the band along and watch what it does at that dot.
In those places a family that added a fourth and fifth child in the affordable early years meets the true cost of those children a decade later, when two or three teenagers can be away at dormitory yeshivas at the same time. The step was never missing. It was waiting, further down the calendar than the birth that created it, and it belongs to the community that had no seat to offer a teenager rather than to out-of-town life as a whole.
Does a tall step actually cost a family a child?
This is the human question under all the arithmetic, and it deserves care. We cannot prove that a tall step prevents a birth, because the family that chose a modern Orthodox town also chose its bedrooms and its schools, and those choices cannot be cleanly separated. What we can show, in a deliberately simple model, is that the parity where each community's step is tallest is the parity where the decision to go on grows hardest.
Each line in the chart below is one kind of family. The three positions along the bottom are three decisions: whether to go on to a third child, then to a fourth, then to a fifth. Height is how likely that family is to go on, read against the scale down the left side, so a line that falls between two positions belongs to a family growing less likely to have the next child. Colour marks the community, red for modern Orthodox, teal for yeshivish, gold for out-of-town. A dashed line is the harder version of its community: kollel rather than working, no local high school rather than one.
Follow each line left to right and find the ones that go down.
The modern Orthodox line dips exactly at the fourth child, then recovers. The out-of-town line without a high school sinks at the fifth, where the dormitory bill comes into view. The yeshivish lines never dip, because that staircase never builds a wall. Its story is the height of the whole climb against a modest income, which is a different kind of hard.
The economics can cost a family one more child. Which child it costs is written by where that community put its tallest step.
So: in a large Orthodox family, which child is the expensive one?
Whichever one the community priced highest. In a modern Orthodox town it is the fourth, and the fifth costs a great deal less. In a yeshivish town it is the third, and every child after that is cheaper than the one before. Out of town it may not be a child at all yet, but a ninth grader who was born five years before the bill shows up.
It would be easy to read all this as a ledger of burden. It is closer to the opposite. Each of these staircases was built by a community trying to raise its children well. The modern Orthodox wall is made of good schools and homes where each child has space. The yeshivish handrails are made of sibling discounts and a culture that shares a bedroom without complaint. The out-of-town bargain, and its delayed bill, are made of a community small enough to be affordable and, sometimes, too small to hold a high school.
No family in any of these worlds is doing the math wrong. They are climbing a staircase their own community poured, and where they pause is not a failure of will. It is the shape of the step in front of them.
Which is also where the hope lives. A step is a thing you can lower. Bedroom-scaled tuition relief and family caps flatten the modern Orthodox wall. A firmer income floor eases the yeshivish climb. One more local high school erases the out-of-town delay. The finding is not that large families are unaffordable. It is that affordability is a design choice, made one step at a time, and it can be made differently.
The fourth child is never too expensive in the abstract. It is only ever as expensive as the step we chose to build under it.
Every dollar figure is a steady-state annual marginal cost, computed from the parameters in Heshbon's Community Database through the project's all-in cost model. Housing uses a bedroom price ladder (a four-bedroom home costs 1.28 times a three-bedroom, a five-bedroom 1.62 times) on 2026 Zillow anchors, with a thirty-year mortgage at 6.6 percent. Tuition is a lifecycle-blended per-child figure, net of a 15 percent aid discount, adjusted by sibling discounts and, out of town, by the Tennessee ESA voucher.
The three communities are stylized structures calibrated to Lakewood, Teaneck, and Memphis-type anchors, chosen because they span the parameters that drive the steps. They describe a way of living, not any one household. Real families vary widely inside each.
The room-sharing norm decides where the housing step fires: at two or more children per room the modern Orthodox fourth-child wall disappears and that child falls from about $38,800 to about $21,800. The depth of the sibling discount decides how gently the yeshivish staircase descends. Both are the author's assumptions and are examined in the working paper's sensitivity analysis, not measured facts.
The proceed-probability model is illustrative, chosen to show a mechanism, and is not estimated from birth records. Because families select into the very community norms that build the steps, cost accounting alone cannot identify a causal fertility effect. A real test would need day-school sibling-progression records, the enrollment-history build flagged elsewhere in this project.
The paper behind this story
Tell me who you are and where you work. I’ll send the complete paper.
Sent to the author directly. No list, no forwarding, no third parties.