Migration is a lot like rain. When we receive a lot of migration (or rain), having prepared and invested for it, and directed it into the most needed areas, it can make us all better off. Conversely, receiving more migration (or rain) than we are ready for, in the wrong places, can make things worse. Knowing it could rain but never buying a raincoat seems to be an unfortunately good analogy for how we currently seem to approach migration.

At present New Zealand is receiving extremely high levels of net migration, but this is like a cloud burst – migration is expected to fall back to ‘normal’ within the next five years, and comes after a several year-long migration drought during the pandemic.

Much like rain – we often think about what’s happening to net migration over the next year or two, akin to the weather forecast. This forecast can drive reactionary policies, such as seeking to restrict migration, when we are at times of high migration. However, we need an eye to the future, thinking about the long-term demographic factors influencing migration and our need for migration, in the same way that we think about what our warming climate will mean for weather in the future.

In this article we discuss the longer-term outlook for New Zealand’s population, to provide some balance to our often focus on near term population growth. We look at how the drivers of population growth are changing, where our population might land in the year 2100, and what might happen if we planned migration to limit population growth.

Over the past 70 years, population growth has been in abundance in New Zealand. We were a young and growing population, not to mention attractive to migrants. Through the 1950 and early 1960s, strong birth rates meant that New Zealand’s population grew by an average of 1.6%pa from natural increase alone, with net migration adding an average of 0.4%pa on top of the natural increase, for total growth of 2.0%pa (see Chart 1).

Birth rates started to fall later in the 1960s into 1970s, leading to weakening natural increase and population growth overall. Fast forward to the 2020s, and fertility rates continue to ease while the childbearing population is flat. What’s more, the population born in the fertility surge of the 1950s are entering old age, leading to increases in deaths. Over time, net migration is becoming increasingly significant to overall population growth, as the rate of natural increase recedes to below 0.5%pa – a consistent decelerating trend which has been playing out for decades.

A growing population and easing natural increase means that looking at just the pure level of net migration in isolation doesn’t tell the whole story. In 2002, international net migration spiked to 51,200, pushing overall population growth to 2.0%. However, if we received 51,200 net migrants in 2023, that would have only pushed overall population growth to 1.4%. If we receive 51,200 net migrants in 2050, our population will only grow 1.0% given our larger population. And the ever-increasing share of net migration being used to offset the ageing of our population means that an increasingly smaller share of net migration is actually growing our population.

Over time, our natural increase (more births than deaths) will switch to natural decline (more deaths than births). Our population future will increasingly be dependent on net migration just to maintain the current population, and will be the only way we can increase the population in the medium-term. To illustrate this sensitivity, we’ve projected New Zealand’s population in the year 2100 based on nine different levels of net migration from 10,000 to 90,000 per year. Underlying this we’ve assumed net migration to 2028 will follow our current macroeconomic forecast and used Stats NZ’s medium projection for fertility and mortality throughout.

If net migration averages 10,000 per year, our population in 2100 would be 5.7m, barely above the 5.2m today (see Chart 2). Our medium projection of 30,000 annual net migration would bring the population to 7.7m in 2100. Considering a higher level of net migration, our population in 2100 could range from 8.7m (based on 40,000 net migrants per year) up to 13.8m (based on 90,000 net migrants per year).

Global competition for migrants is likely to heat up over the next decade, with an ageing population across the developed world and slowing population growth across the developing world too. A net migration level of 90,000 per year would appear to be exceptional in that context, but clearly there is a range of outcomes for New Zealand depending on how permissive we want to be towards migrants.

We can and do exercise a great deal of control over the migration of non-citizens to New Zealand, as anyone who has dealt with our immigration system can attest. However, the international migration of New Zealand citizens is a significant factor in overall population growth. With effectively no ability to restrict the migration of citizens, we can’t control this component (once-in-a-lifetime global pandemics being an exception – and even then, pathways through MIQ were required). This inability to directly influence NZ citizen movements is especially important given New Zealand is estimated to have the third largest diaspora populations in the OECD, with an estimated 600,000 New Zealand born people living in other OECD countries globally – over three quarters of whom live in Australia. Australians are also welcome to live and work here indefinitely, so represent another aspect of our migration which can’t be controlled (without upsetting trans-Tasman relations, at least).

The most powerful lever on population growth is the approval of non-citizens to arrive in New Zealand through the immigration system. Over the past twenty years, annual arrivals of non-New Zealand or Australian citizens has ranged from a low of 20,300 in 2021 to a high of 178,300 in 2023, generally siting in the 60-100,000 level. Net migration of non-New Zealand or Australian citizens has ranged from -20,200 in 2021 to 143,900 in 2023, generally ranging between 20-60,000.

To illustrate how a limit on non-New Zealand or Australian migration would play out, we’ve estimated New Zealand’s population in 2023 as if different migration limits or maximums had been implemented over the past 20 years. We’ve done this based on limiting net migration, because departures and arrivals are closely correlated, as many people leave a year or two after arriving. In practice, a migration limit would likely apply to the number of arrivals (because the government generally can’t as easily influence the number of migrants leaving New Zealand).

Chart 3 shows that nearly any limit would have restrained migration in 2023, but it would take a considerably lower limit to constrain migration over the past twenty years. This lower limit reflects that such a limit would only be binding in a handful of years when migration is high. A 160,000pa non-New Zealand or Australian arrival limit would have only been binding in 2023. Migration limits ranging from 80,000pa and 160,000pa would all have seen average population growth at 1.2%.

A migration limit that was binding over 2014-2016, and 2023-2024, would have helped by buying us time to develop sufficient housing and infrastructure for extra population growth. However, such a limit would risk exacerbating labour shortages. It’s crucial that we don’t swing the pendulum too hard the other way – sending a message that we don’t want migration. Over the long term, New Zealand will most certainly need migration just to keep the lights on, let alone grow.

This simplistic analysis doesn’t take into account various feedback effects. Limiting non-New Zealand and Australian arrivals could lead to skills shortages in some areas, which might be solved by higher migration of New Zealand and Australian citizens, or lower departures of non-citizens. Limiting migration over the past 20 years would have constrained growth in our childbearing population, so we would have fewer children today if we had limited migration.

Having and working towards a defined population target would be incredibly useful, almost irrespective of what the actual target is. Our population growth is now highly sensitive to net migration, and spikes in net migration can apply considerable pressure. Working towards a target by strategically planning for migration could smooth out the spikes and reduce uncertainty around our growth trajectory. This reduced uncertainty would in turn reduce (but not eliminate) the amount of guesswork across central and local government in planning for the future. The need for a target or range has long been raised, and came through clearly in the Productivity Commission’s analysis of migration settings.

Let’s have a discussion of where we want to head as a country and build cross-party consensus on a population target. If we fit an extra 2 million people in the golden triangle of Auckland, Hamilton, and Tauranga, they would together form a mega-city of a similar size to Sydney. Alternatively, we could aim to decrease our population as we retreat from low-lying coastal areas, and avoid expanding our urban footprint.

Most importantly, we should use a population target to plan assertively for that population, with more consistent up-front investment into infrastructure ahead of time to ensure that the rapid appearance of high population growth (more demand) is serviced by enough infrastructure capacity (supply), instead of current settings. Right now, high population growth will eventually spur further investment that will occur too late and make the population incur higher costs in the meantime, as demand outstrips supply.

A population target doesn’t need to be defined down to the last person, but an idea of where the population is heading means we could at least plan to meet the needs of that population. So that when the forecast is issued, and the rain comes, we’re not left outside soaking wet and wondering who to blame.