America's Population Was Going to Crash, Then Explode, Then Crash Again: A Century of Demographic Certainty That Kept Being Wrong

Demography is the science of counting people and predicting what they will do next. It is, by the standards of the social sciences, admirably empirical. It is also, by the standards of actual history, reliably humbled by the people it is counting, who persistently decline to behave according to the projections.

America's population has been in demographic crisis approximately once per decade since the 1920s. The nature of the crisis changes. The certainty with which experts describe it does not.

The Empty Cradle Panic of the Early Twentieth Century

By the 1910s and 1920s, a significant portion of the American intellectual establishment had convinced itself that the country's birth rate was heading for catastrophic collapse. The concern was not evenly distributed—it was focused, with anxieties that were inseparable from the era's uglier racial preoccupations, on the declining birth rates of native-born, white, Protestant Americans relative to immigrant populations from Southern and Eastern Europe.

This was the milieu that produced the Immigration Act of 1924, which imposed strict national-origin quotas designed, in part, to manage the demographic composition of future Americans. The predictions driving this legislation were confident: without intervention, the ethnic character of the republic would be transformed within a generation.

What actually happened: the immigration restrictions held for forty years, and then the Immigration and Nationality Act of 1965 dismantled the quota system—producing demographic changes that the 1924 restrictionists had predicted, just on a longer timeline and for reasons that had nothing to do with their original models. The country was transformed. It was also, by most measures, fine.

The Baby Boom Nobody Predicted, Followed by the Bust Nobody Expected

If you want a masterclass in demographic forecasting failure, study the years between 1945 and 1975.

After World War II, demographers expected birth rates to return to the pre-war downward trend. The Depression had suppressed family formation. Women had entered the workforce. The consensus forecast was for continued slow population growth, maybe stabilization. What actually happened was the Baby Boom: a sustained, dramatic surge in birth rates that lasted from 1946 to 1964, producing roughly 76 million Americans and an entire cultural infrastructure—suburbs, schools, pediatricians, station wagons—built to accommodate them.

Nearly nobody saw it coming. The models were wrong. The people, apparently, had other plans.

Then, having failed to predict the boom, demographers overcorrected magnificently for the bust. By the late 1960s and early 1970s, birth rates were falling sharply, and the field pivoted with impressive speed from worrying about too few babies to worrying about too many people.

Paul Ehrlich's The Population Bomb, published in 1968, predicted mass starvation and civilizational collapse driven by overpopulation. It sold three million copies. It was wrong in ways that are instructive: the Green Revolution dramatically increased agricultural yields, birth rates in developed countries continued falling, and the apocalyptic timeline proved, as apocalyptic timelines often do, to be optimistic about the pace of catastrophe and wrong about its nature.

The U.S. fertility rate fell below the replacement level of 2.1 children per woman in 1972. The empty cradle panic, which had briefly been replaced by the overcrowded planet panic, quietly returned.

The Immigration Wild Card That Kept Reshuffling the Deck

Any demographic forecast about America that does not adequately account for immigration is, historically speaking, going to be wrong. This has not stopped each generation from making exactly that mistake.

The 1970s and early 1980s produced a wave of projections about a graying, shrinking America—a society struggling with the fiscal implications of a large elderly population and an insufficient number of young workers to support it. These projections were not unreasonable given the data available. They were also made at the beginning of one of the largest sustained immigration waves in American history.

Latin American immigration, which accelerated dramatically through the 1980s and 1990s, and Asian immigration, which grew steadily across the same period, reshaped the demographic picture in ways that the forecasters of 1975 had not modeled. The U.S. population, which some projections had plateauing around 250 million by 2000, hit 281 million in that year's census and kept climbing. The country did not so much defy the predictions as import its way around them.

This is not a criticism of immigrants. This is a criticism of forecasters who kept underestimating the degree to which America's population trajectory is a function of global conditions—economic crises, political instability, climate, conflict—that are, by definition, difficult to predict.

The Current Panic, Which Is Both Old and New

The 2020s have delivered a fresh iteration of demographic alarm, this time with higher production values and more data visualization. Birth rates have fallen to historic lows. Immigration has become the most politically charged subject in American life. Economists warn about labor shortages. Commentators warn about cultural transformation. Everyone is certain about what is going to happen next.

Some of the concern is legitimate—genuinely low birth rates do create genuine fiscal pressures on entitlement programs designed around different population assumptions. This is a real problem worth taking seriously.

What is less clear is whether the current moment of demographic certainty will prove more accurate than the previous ones. The forecasters of the 1920s were certain. The forecasters of the 1960s were certain. The forecasters of the 1980s were certain. Each set of certainties was overtaken by human behavior that declined to read the projections.

Birth rates have surprised before—in both directions. Immigration levels respond to forces that no domestic policy fully controls. The country that was supposed to be emptying in 1975 added 100 million people in the next fifty years.

The Consistent Variable

The one thing demographic forecasters have consistently underestimated is the degree to which Americans—and the people who become Americans—make decisions based on conditions that cannot be modeled from the present. Economic opportunity, political upheaval, medical advances, cultural shifts: these are the actual engines of population change, and they are, almost by definition, the things forecasters know least about.

The country keeps arriving at a future that looks different from the one everyone was certain was coming. This has been true for a century. It will probably continue to be true, which is itself a prediction, and therefore possibly wrong.

We'll update this article when the 2030 census comes back. It will surprise someone.