Note on 'Population Shock', by Abul Rizvi

My notes on: Rizvi, Abul. (2021). Population Shock.

(This note may contain errors and inaccuracies. It was mostly written just for myself, in preparation for my live podcast conversation with Abul Rizvi.)

At just 86 small pages, this is a short book, more an essay. Published in 2021, it can be read as a popularised summary of the author's PhD thesis.

The author, Abul Rizvi, completed a PhD in population and immigration policy at the University of Melbourne in 2020. Prior to that, he worked in the Australian Public Service; he was Deputy Secretary of the Department of Immigration, and managed Australia's migration program from 1995 to 2007.

Abul was intimately involved—indeed, probably one of the top two senior public servants involved—in motivating and designing the Howard Government's 2001 changes to Australia's immigration program. These changes initiated the largest sustained increase in immigration since the "populate or perish" policy of Australia's post-war immigration program.

Modern Australia would be unrecognisable without these changes. Today, thirty per cent of Australians were born overseas—a proportion higher than that of any other major developed country. As the 2023 Parkinson Review noted, despite Australia comprising only 0.3 per cent of the world’s population, 2.8 per cent of the world’s immigrants live here.

Apart from teaching me about the general history of the 2001 changes, two other things in this book stuck out to me in particular.

(1) The 2001 changes to the migration system were a stroke of Australian policy genius.

In the late 1990s, the Australian government realised that it needed to massively increase the level of immigration. This was for two reasons. First, to fill skills shortages. Second, to slow Australia's rate of population ageing. (It's not clear to me which was the more important rationale, though my sense is the second.)

The government wanted to achieve these goals without loosening the selection criteria for skilled permanent migrants. But the problem was that there wasn't a depth of excess applicants.

The solution (described by Abul as "the only workable approach") was to massively increase the temporary migration program, mainly in the form of international students and working holiday makers. Once they’d arrived in the country and upskilled, they could apply for permanent residency. Crucially—and counter to prevailing wisdom—they could do this without having to go back home.

These policy changes, introduced on the 1st of July 2001, achieved their intended result. Since then, more than 2 million international students and working holiday makers have come to Australia, with many of them going on to settle permanently. In recent years, international students have made up 40-50% of net migration (which accounts for both permanent and temporary migrants). It's not too great a simplification to say that in Australia today, immigration policy boils down to decisions about international students.

The 2001 changes unlocked a number of benefits simultaneously:

• More migrants to fill skills shortages, especially for government-funded sectors like health and aged care where government can't or won't increase wages.
• Fiscal benefits from young, skilled migrants whose education costs had already been borne by another country and who were now beginning their high-earning, tax-paying careers in Australia.
• A slower rate of population ageing thanks to migrants skewing younger (and also, by that fact, tending to be in the child-bearing age range).
• A booming tertiary education sector, which quickly grew to become Australia's third largest export, raising living standards for all Australians. (It's not clear whether this was an intended benefit nor whether the Howard Government foresaw the full extent to which education exports would grow.)

(2) The second thing that struck me about Population Shock was Abul's attitude toward fertility declines.

He can't be accused of insouciance—the policy changes he worked so hard to implement were largely motivated by the demographic reality imposed by fertility declines. But he does seem fatalistic, and in a way that I found strangely incurious. Allow me to explain.

Australia’s total fertility rate (TFR) first fell below replacement level (which is a bit above two kids per woman) in 1976. It currently sits at 1.5—the lowest it's ever been.

Such a low fertility rate ensures that as the baby boomer generation travels, like a pig in a python, up the population pyramid, Australia's population will "age". That is, the proportion of people older than, say, 65 will increase. This trend is unfolding across the entire developed world, and in China and Russia.

It poses a monstrous challenge, since each worker must support (through debt and taxes) more and more retirees.

But the story doesn't resolve once the baby boomer cohort passes on. With low fertility rates, population ageing resumes.

Left to run its course, population ageing leads to population shrinking, as deaths begin to exceed births.

And population shrinking, for reasons I'll briefly touch on below, is a very bad situation.

This is where migration becomes a useful tool.

Migrants slow the rate of population ageing in a couple of ways. First, as Australia's migrants skew young, they increase the proportion of people under 65. Second, as Australia's migrant women tend to be of child-bearing age, they also increase the number of births, adding more, even younger, people to the population.

According to Abul, if Australia continues with its current high level of net migration, we won't cross the threshold beyond which deaths exceed births ("natural decrease") until "possibly well into the second half of this century".

If we cut net migration to a very low level, we'd cross the threshold to natural decrease within 20 to 30 years.

And if we dropped net migration to zero, we'd cross the threshold within 15 years. In this scenario, population shrinking would begin immediately upon commencing natural decrease, since population growth is a combination of both natural increase/decrease and net migration.

(In the first two scenarios, I'm not sure how soon after natural decrease begins that population shrinking would begin, since the timing would depend on how strongly net migration offsets natural decrease. You could also have a scenario of stable population size and no growth.)

It's crucial to note that, practically speaking, migration can only slow the rate of population ageing, not put it in reverse. In other words, net migration has diminishing returns for slowing the rate of population ageing. This is a somewhat technical point, but the basic intuition (as I understand it) is that today's immigrants (assumed to have a similar fertility rate to the locals) get added to the population and themselves get older, contributing to a larger ageing population in future. And a larger ageing population means an even larger group of immigrants would be required next time around in order to slow the rate of population ageing by the same amount.

This leads to two conclusions:

1. The proper way to think about immigration as a demographic tool is not as a substitute for natural increase but as a way to buy time in which to deal with the problems caused by natural decrease.
2. If you care about population growth, you ultimately have to address low fertility (or else pray for AGI).

And we should care about population growth.

On page 42, Abul mentions in passing:

"Once a nation’s population is severely aged and in ongoing decline, the prospect of a ‘technical recession’—that is, two consecutive quarters of negative economic growth—will become more common, as has frequently been the case in Japan. As more nations move into the fourth stage of the population shock, economists will need a new definition for economic recession. The current definition was born out of the second stage and will be of limited use as developed nations move further into the third and fourth stages when negative economic growth becomes more common."

But wait a minute. This is literally a dystopian scenario.

In endogenous growth models, à la Paul Romer, the economic growth rate is tied to the population level. In semi-endogenous growth models, à la Chad Jones, the economic growth rate is tied to the population growth rate. Whichever story you buy, a shrinking population implies slowing or falling economic growth. And needless to say, as the size of the overall economic pie stalls or shrinks expect more people to fight over the size of their slice, expect cooperation to drop to lower scales, expect more violence.

So we shouldn't accept low fertility as a fate imposed by the universe. Or at least, we should be intensely curious about its causes.

Abul believes that fertility declines have two major causes: (i) the introduction of contraceptive technologies (principally the pill) and (ii) increasing female education and workforce participation. These changes or trends are obviously not things we'd want to stop or wind back even if we could, and so we're stuck in a low fertility world.

I'm not sure why Abul is confident that contraception is so important. Contraceptive technologies no doubt make it easier to have fewer kids, and may have allowed fertility to fall a little further than it otherwise would have. But the demographic transition predates contraception—it begins first in France, by about the 1830s, and then in other parts of Europe, including Britain and Germany, by about the 1880s.^[1]

The contraception story projects modern cultural attitudes backwards in time and assumes people in the past would have preferred, like we moderns, to control and curtail their family sizes but were as sex-crazed as us and, lacking contraception, could do nothing but grudgingly accept all the kids that that would entail. This story doesn't permit that our ancestors might in fact have had radically different views about family size.^[2]

The women's empowerment story feels closer to an important causal factor, but in my view is still incomplete: it may, at least in part, be epiphenomenal to fertility declines. Fertility declines could be better explained by a deeper cultural story, namely that we just no longer view having lots of kids as a very prestigious thing to do.

When the family unit is the bastion of culturally-inherited beliefs, family-size-limiting beliefs struggle to gain currency, since the family lineages that hold and transmit them will eventually be filtered out by natural selection. But in a world in which school, media and other social institutions also transmit culture, cultural norms that limit family size can escape natural selection and proliferate. Of course, such "non-parental modes of cultural transmission" are part and parcel of the broader process of modernisation, making causal inference very hard indeed.

Last year, I did a modest amount of reading on fertility declines and came away feeling very confused but convinced it was one of the most important questions to be answered. While fertility declines feel overdetermined, the single best framework I've found so far for thinking about the phenomenon is the cultural evolution one I've been evincing, which was put forward by Rob Boyd and Pete Richerson.