A falling birth rate is a conservative problem, not a culture-war slogan

Every pension actuary in the country knows something that most politicians prefer not to say aloud: a retirement system built on the young paying for the old is a bet on babies. You can call it insurance, you can call it a trust fund, you can print a card with your name on it. Underneath the label, the money that arrives in a retiree's mailbox this month was earned by a worker this month. The only way the arrangement lasts is if there continue to be workers.

So here is a fact that ought to interest anyone who cares about the budget, the family or the future: American women are having fewer children than are needed to replace the population. That is not a slogan. It is arithmetic, and it has consequences for both columns of the ledger, the fiscal and the moral. This post defines the terms, shows the numbers I can stand behind, states the argument in four steps, and then lays out the conservative disagreement over what to do, including the best libertarian objection.

Defining the terms

The total fertility rate, which the National Center for Health Statistics (NCHS) publishes each year in its vital statistics reports, is the average number of children a woman would have over her lifetime if she experienced the birth rates of a given year at every age. It is a snapshot, not a forecast, and it removes the distortion that comes from a population having more or fewer women of childbearing age.

The replacement level is the fertility rate at which a population exactly replaces itself from one generation to the next, absent migration. In a developed country with low child mortality it is about 2.1 children per woman. The extra one tenth above two accounts for children who die before reaching adulthood and for the fact that slightly more boys than girls are born.

A pay-as-you-go system is one in which current taxes pay current benefits, with no large fund set aside in advance. Social Security's retirement benefit is largely of this kind. The system has a trust fund, and what the fund holds is considered in What the Social Security and Medicare trust funds hold, and depletion day, but the flow of payroll taxes from today's workers covers most of what goes out today.

What the numbers show

Here are the facts I am certain of, with the instruction to check the latest from the source.

American fertility was above replacement for much of the postwar era. After a famous peak in the late 1950s, which saw rates well above three children per woman, it fell in the 1960s and early 1970s and dipped below replacement in the 1970s. It rose again toward replacement in the late 1980s and the 2000s. By the NCHS's own account, the rate has been under the 2.1 replacement level every year since 1971 except 2006 and 2007, and in recent years it has been well below. The country has therefore been below replacement continuously since 2008, the better part of two decades. You can read the year-by-year series in the NCHS "Births: Final Data" reports or in the National Vital Statistics Reports, which is where I would send you rather than quote a decimal point from memory.

The second set of numbers is in the annual report of the Social Security Trustees. One table in that report gives the number of covered workers for each beneficiary. In 1960, there were about 5.1 workers for every person receiving Social Security benefits. By the time of recent reports, the ratio has fallen to under three, and the Trustees project it falling further. Read the table yourself in the Trustees' report; they have published the series for decades.

The arithmetic of the ratio

The workers-per-beneficiary ratio matters because of what it does to the burden on each worker. Let me show it with a simplified example. It is a hypothetical: it treats the system as purely pay-as-you-go and ignores the trust fund, the wage cap and benefit differences, so it illustrates the principle and does not predict any actual tax rate.

Suppose each beneficiary receives a benefit equal to some share of the average wage. If 5.1 workers support each beneficiary, then each worker carries one part in 5.1 of one benefit. One divided by 5.1 is about 0.196, or 19.6 percent of a benefit per worker. Now suppose the ratio falls to 2.7. One divided by 2.7 is about 0.370, or 37.0 percent of a benefit per worker. The same benefit, spread over fewer workers, takes nearly twice as large a bite out of each paycheque. That is the whole problem in a single sentence: if the benefit stays the same and the number of workers per beneficiary falls, the tax per worker must rise, or the benefit must be cut, or the difference must be borrowed.

The payroll tax that funds Social Security is 12.4 percent of wages up to a cap, split evenly between employer and employee at 6.2 percent each, and Medicare hospital insurance adds 2.9 percent, also split. Those are the legislated rates, and any change in the ratio shows up as a gap that Congress has to close.

Falling fertility is not the only reason the ratio fell. People are living longer, which is good news, and the large postwar generation is retiring. But fertility is the one factor that determines how many workers there will be in thirty years, because they have to be born first. Immigration can offset a shortfall, and I take up the immigration side in The Case for Nuclear-Family Immigration Over Extended Family Chains and its neighbours; the point here is that a country cannot responsibly plan a pension system on the assumption that other countries will keep sending it workers while it produces few of its own.

That ties to the fiscal picture in general. A shrinking workforce supporting a growing retired population raises the pressure on the deficit, a pressure described in Seven myths about the national debt that Washington finds convenient and visible on the national debt clock.

The argument in four steps

Let me put the claim as a point-by-point sequence of premises.

One. A population that has fewer children than replacement will, absent offsetting migration, shrink and age.

Two. Entitlement programs that promise fixed benefits financed by current payroll taxes depend on a stable or growing ratio of workers to retirees.

Three. Therefore sustained below-replacement fertility raises the cost of those programs per worker, which means higher taxes, lower benefits or more debt.

Four. Beyond the fiscal, a country in which people want children and cannot afford or find the conditions to have them has a problem that no budget fixes. Survey after survey, as I understand it, finds that Americans say they want more children than they have. I will not cite a number, but if that gap exists, it points to obstacles, such as the cost of housing, the delay of marriage and the weakness of the institutions that support young families, rather than to a change in desire.

The first three steps are arithmetic. The fourth is a judgement about values, and I will own it: a society that finds it difficult to form families has lost something that a growth rate cannot measure.

The conservative split

Conservatives agree on the diagnosis far more than they agree on the cure. Here are the two main camps.

The pro-natal policy camp says the state should reduce the cost of having children through tax and benefit design: a larger child tax credit, bigger dependent deductions, one-time payments at birth, help with childcare and housing. The logic is the same as any other incentive. If children cost more than parents can bear, lower the price. I compare these tools in baby bonuses, tax credits or cheaper houses, comparing pro-family policies, and I would encourage you to read the table there. The weakness of this camp is the evidence: studies of cash incentives in other countries suggest small effects on the number of births, though they may shift timing. And a subsidy is a subsidy, paid for by taxpayers, including those who have no children.

The cultural-renewal camp says that money is not the main obstacle. People have children when they marry, when they have stable work, when they belong to a community, and when they believe that raising a family is worth the sacrifice. The causes of decline, on this view, are a later and less stable marriage pattern, a culture that treats children as a lifestyle choice rather than a calling, and institutions such as churches and schools that have lost their hold. The remedy is persuasion, example and the rebuilding of those institutions, which is the territory of 'The success sequence is just correlation,' answered point by point and the family came before the state, so why does the state act like a parent?. The weakness of this camp is that culture moves slowly and politicians are not good at changing it.

A sensible conservative can hold both views at once: use the tax code to stop penalising families, as argued in the marriage penalty buried in welfare programs, with the math shown, and use the pulpit and the dinner table to say that children are a good.

The libertarian objection

Now the best case from the other side of the conservative coalition, stated as its defenders would state it.

The objection says that the number of children a couple has is none of the state's business. A government that sets out to raise the birth rate is a government that treats citizens as a resource to be managed, and that is a worse problem than a shrinking pension fund. If people decide to have fewer children, they are weighing their own circumstances and exercising their own freedom, and a free society honours the result. Pro-natal subsidies are a form of central planning, and the history of such schemes, in countries that tried hard to raise birth rates, is not encouraging. If entitlement programs are unsustainable, the solution is to reform the programs, by raising the retirement age, changing the benefit formula or moving toward personal accounts, and not to ask citizens to produce workers for the system.

That is a serious argument, and it is half right. The state should not decide how many children a family has, and any policy that hints at quotas deserves rejection. A government that wants more babies and lacks the humility to know it cannot order them is a government to be watched.

But the objection proves less than it claims. First, a family policy that lowers the burden on parents does not tell anyone how many children to have; it reduces a penalty. The tax code already shapes family decisions, as the marriage-penalty arithmetic shows. The question is not whether the state influences family choices but whether it does so deliberately and fairly. Second, the libertarian remedy of reforming entitlements is correct and should be pursued, and it does not conflict with the pro-family agenda. Both can be done. Third, the objection treats children as a private consumption good, like a boat. Children are also, in the plain economic sense, the next generation of taxpayers, neighbours, soldiers and caregivers, and the benefit of raising them is shared by people who never paid for it. When a private choice creates a large shared benefit, economists call it a positive externality, and they generally agree that it is under-supplied without some support. That is a conservative argument too, because it rests on the reciprocity between generations that Burke described as a partnership between the dead, the living and those yet to be born.

The Founding perspective

The Founders thought about population, and they thought about it as a strength. The Constitution's requirement of a decennial census in Article I, Section 2, is a recognition that a republic's representation depends on its people, and that the number and distribution of citizens is a public concern. In Federalist No. 14, Madison argued that the republic could be extended over a large territory. The assumption behind that vision was a growing nation of families.

Behind the Declaration's "pursuit of Happiness" lies a society of households that expected to produce the next generation. A people that does not renew itself is a people that has stopped believing in its own future. That is the moral point behind the fiscal one, and it is the reason I regard a falling birth rate as a conservative concern: conservatism is the philosophy of inheritance, and inheritance needs heirs.

What a sensible response looks like

Here, briefly, is a program that a conservative of either camp could support.

1. Stop punishing families in the code. Remove the marriage penalties and widen the family tax provisions.
2. Make housing for families possible. Zoning that bars smaller lots and starter homes is a tax on young families. The example of houston has almost no zoning, and its houses cost less to prove it is the best natural experiment we have.
3. Reform the entitlements honestly. Do not wait for the ratio to force a crisis.
4. Rebuild the institutions that sustain families, from churches to schools, rather than leaving the work to government programs.
5. Treat children as a good in the public conversation. A culture that speaks of them only as a cost will get fewer of them.

The case in brief

1. American fertility has been below the 2.1 replacement level for most of two decades, according to NCHS data, and the Social Security Trustees report that the number of workers per beneficiary has fallen from about 5.1 in 1960 to under three.
2. In a pay-as-you-go system, fewer workers per beneficiary means a higher burden per worker: in the hypothetical, 1 divided by 5.1 is 19.6 percent, while 1 divided by 2.7 is 37.0 percent.
3. Conservatives divide between pro-natal tax and benefit policy and cultural renewal, and the sound answer uses both.
4. The libertarian objection is right that the state must not manage family size, but wrong to treat children as a purely private good when their benefits are widely shared.
5. A republic that does not renew itself has stopped believing in its own future, which is why the birth rate is a conservative concern.

What you can do: open the latest Social Security Trustees report, find the table of covered workers per beneficiary, and ask your representative through the representatives page how Congress plans to close the gap.