Baby booms and busts: how population growth spurts affect the economy

A baby boom is generally considered to be a sustained increase and then decrease in the birth rate. The United States, the UK and other industrialized economies have experienced only one such baby boom since 1900 – the one that occurred after World War II.

In addition, many currently developing economies such as India, Pakistan and Thailand have experienced a baby boom since 1950 as a result of a sustained decline in infant and child mortality rates as a result of improved medicine and sanitation.

So what’s the economic impact of these baby booms? Do demographics play a role in determining when an economy expands and contracts? Do they boost incomes or cause them to fall as more young people enter the workforce? I’ve been studying the impact of baby booms on wages, unemployment, patterns of retirement and gross domestic product (GDP) growth for 20 years and, while there are some questions that haven’t been answered, here’s what we’ve learned so far.

Negative impact on employment

The initial impact of a baby boom is decidedly negative for personal incomes.

Baby booms inevitably lead to changes in the relative size of various age cohorts – that is, a rise in the ratio of younger to older adults – a phenomenon first described by economist Richard Easterlin. (In statistics, a cohort is a group of subjects who have shared a particular event together during a particular time span.)

These effects cause a decline in young males’ income relative to workers in their prime, a higher unemployment rate, a lower labor force participation rate and a lower college wage premium among these younger workers.

This occurs because younger workers are generally poor substitutes for older ones, so the increased supply of youths leads to these negative employment results.

Back in the 1950s, entry-level young males in the US were able to achieve incomes equal to their fathers' current income. This was because of that age group’s reduced relative size as a result of the low birth rates in the 1930s. But by 1985 – about the time the peak of the baby boom had entered the labor force – that relative income had fallen to 0.3; in other words, entry-level men were earning less than one-third of what their fathers made.

In developing countries, these relative cohort size effects – the reduction in young males' relative income and increase in their unemployment rate – are multiplied by the impact of increasing modern development, especially the rising level of women’s education.

In addition, the large influx of baby boomers into the labor market in the US forced many older workers, who would otherwise be working in “bridge jobs” prior to retirement, into earlier retirement. This explains how the average age of retirement for men and women went down in the 1980s.

This decline in income relative to their parents and their own material aspirations has a host of repercussions on family life. It leads to reduced or delayed marriage, lower fertility rates and increased female labor force participation rates as young people struggle to respond to their worsened prospects.

From boom to bust … to boom?

The reduction in relative income – which the US experienced in the ‘60s and '70s – thus results in a subsequent “baby bust” as people delay starting a family.

It was hypothesized that these baby booms might be self-replicating as reduced birth rates on the trailing edge of the boom caused the subsequent cohort to be smaller in size, thus leading to better labor force conditions, increased birth rates and an “echo boom” in the next generation.

This theory was based on what led to the baby boom in the first place, when the favorable labor market conditions experienced in the 1950s emerged as a result of fewer children being born during the 1930s, reducing the young-to-old-adult ratio.

Though the echo boom of the 2000s represented an increase in the absolute number of young adults, it didn’t lift their cohort size relative to their parents because birth rates have remained fairly stable at low rates since the end of the post-WWII baby boom.

That has not, however, translated into significantly better labor conditions, at least not the kind experienced by young adults in the 1950s that led to the baby boom. The reasons for this phenomenon have not yet been explained.

So can changing demographics cause recessions?

Another way of exploring the effects of changes in the proportion of young adults in the population is to look at fluctuations in the relative size of the young adult population over time. These seem to have a significant effect on the economy.

As young adults move out of high school and college and set up their own households, they generate new demands for housing, consumer appliances, cars and all the other goods attendant on starting a new adult life. These new households don’t account for a large share of total expenditures, but they represent a major share of the growth in total consumer expenditures each year.

So what happens if, after a period of growth in this age group, the trend reverses? It is likely that industries counting on further strong growth will be forced to cut back on production, and in turn to cut back on deliveries from suppliers – which will in turn cut back on deliveries from their suppliers, creating a snowball effect throughout the economy.

This picture is supported by the patterns over the past 110 years depicted in the graph shown below.

The graph tracks the three-year moving average of the annual rate of change in the proportion of young adults in the US. The red vertical lines indicate the beginnings of recessions. Data past 2020 are projections.US Census Bureau

The curve on the graph represents a three-year moving average of the annual rate of change in the proportion of young adults in the US population, as given by the United States Census Bureau. “Young adults” are defined as those aged 15-19 prior to 1950, and 20-24 in the years after, given changing levels of education over time. This curve is overlaid with vertical lines that mark the start of recessions, as defined by the National Bureau of Economic Research.

There is a very close correspondence between the vertical lines, and peaks in the curve, as well as points where the curve turns negative. In addition, the deep trough between 1937 and 1958 contained another four recessions, and there were two in the trough between 1910 and 1920 (not marked on the graph). The only recessions over the last 110 years that don’t appear to correspond to features of the curve, are those in 1920, 1926 and 1960.

The pattern of causation – if it is one – cannot run from the economy to demographics, since these are young people born over 15 years before each economic downturn. In addition, there’s a one-year lag in the age groups that has been used to control for possible migration effects of a recession – that is, how many people left the US as a result of worse labor market conditions.

The fact that no “double dip” recession occurred in 2012, even as the share of young people fell that year, might be the result of the economic stimulus applied after the most recent recession.

Food for future thought

Obviously there are many other factors associated with economic downturns, but aspects of the empirical regularity demonstrated here can be seen in many countries over the past 50 years – especially regarding the international financial crises of 1980-82, 1992-94, and 1996-98 and 2007-2008.

This is not to say that demographics were the sole cause of the recessions, but rather that they influenced the timing of such events, given a host of other possible factors. For example, did they play a role in determining when the recent housing bubble burst? That question has yet to be answered, but further study may shine some light.

This article is part of a series on What’s next for the baby boomers.

Diane J Macunovich does not work for, consult, own shares in or receive funding from any company or organization that would benefit from this article, and has disclosed no relevant affiliations beyond the academic appointment above.