THE ECONOMICS OF IMMIGRATION

According to Saul Eslake, one of Australia’s leading economic commentators, the closure of Australia’s border during the 2020-21 pandemic has given government policy makers “an unprecedented opportunity to take stock of Australia’s migration policy settings”. Contributing economics editor at The Australian Judith Sloan agrees and adds that “The bottom line is that it’s impossible to presume that immigration will provide a net boost to the budget because the number of skilled primary applicants is such a low proportion of the total migrant intake and many temporary migrants won’t pay any tax at all.” To continue that discussion, Saul Eslake and Judith Sloan presented a Sydney Institute Zoom meeting on Tuesday 3 August 2021.

THE ECONOMICS OF IMMIGRATION

SAUL ESLAKE

This talk is about the economics of immigration – and so, inevitably, it is going to leave aside most of the other dimensions of immigration which are at least as important as the economics is in any assessment of its overall impact. Australia’s above-average rate of population growth has been a significant, but under recognised, contributor to Australia’s “headline” economic performance over the past two decades:

• Over the twenty years to the final quarter of 2019 (that is, before the onset of the pandemic), Australia’s economy grew at an average annual rate of 2.8-0.9 percentage points above the OECD average of 1.9 per cent per annum;
• Over the same period, Australia’s population grew at an average annual rate of 1.5 per cent – more than double the OECD average of 0.7 per cent per annum;
• Stripping out the effects of this substantially above-average population growth rate, Australia’s per capita GDP growth rate was marginally below the OECD average, 1.23 per cent per annum compared with 1.25 per cent per annum for the OECD as a whole.

Had Australia’s population grown at the same rate as the OECD average since the fourth quarter of 1999, and everything else been equal, instead of having had only three quarters of “negative growth” in real GDP over the following 20 years (and none of them in succession), Australia would instead have had five (and we probably wouldn’t have been able to say we’d gone that length of time without a recession, for whatever that boast might have been worth).

58.5 per cent of the increase in Australia’s population over the 20 years to the final quarter of 2019 was directly attributable to net overseas migration. In other words, without the almost 3.9 million more people who came to Australia as some kind of immigrant than who left over this period, Australia’s population would have risen at an average annual rate of 0.7 per cent – that is, at the same rate as the average for OECD member countries.

This calculation probably understates the contribution of migration to Australia’s population and “headline” GDP growth rates, because migrants will themselves have had children during this period.

Let me emphasise that I am not saying that Australia’s immigration program has been a Good Thing because of the contribution it has made to “headline” GDP growth. Like most economists (and many others besides) I do not advocate the pursuit of growth in GDP as an end in itself. Rather, like most economists, I regard growth in per capita GDP as a more appropriate measure of (material) living standards – and therefore, real per capita GDP growth as a more appropriate measures of whether, and of the extent to which, material living standards of living are improving.

I should perhaps also add here that, like most economists, and many others besides, I recognise that per capita GDP growth in isolation from measures of its distribution tells us nothing about the extent to which the material living standards of all the different components of a society are improving – that is, whether a rising tide (if that’s what it is) is in fact lifting all boats.

A more important initial question is, therefore, whether Australia’s immigration has made a positive contribution to per capita GDP.

An investigation by the Productivity Commission published in 2016 concluded that Australia’s immigration program does have a positive impact on per capita GDP. It estimated that continuing Australia’s migration intake at its “long-term historical rate” with the same age and skills composition would boost Australia’s per capita GDP by about 7 per cent (equivalent to around $7,000 per person in 2013-14 dollars, or about $7,950 in today’s dollars) by 2060.

Like most other economists, I also recognise that there are many aspects of both individual and societal “well-being” – things which both enhance and detract from “well-being” in its broadest sense – that are not comprehended by, or included in, GDP, whether expressed in per capita terms or not.

All of those considerations are relevant to any evaluation of the contribution of immigration to Australia’s “well-being”, and that of individual Australians.

Quite properly, therefore, the Productivity Commission then went on to stress that “whether migration delivers an overall benefit to the existing Australian community will also depend on other factors”, including the distribution of the economic benefits, and what it called “the broader impacts of immigration, notably the associated social and environmental impacts”.

The Commission noted that “increasing numbers of immigrants can adversely affect the quality of Australia’s natural and built environment unless governments take action to mitigate congestion and other pressures”. It also noted that “some environmental impacts, such as the recreational value of near-empty beaches, and the value of biodiversity, are hard to measure, let alone monetise”.

Nonetheless, the Productivity Commission concluded that “overall, some positive rate of immigration within Australia’s absorptive capacity is likely to deliver net benefits to the Australian community over the long term”.

That general conclusion is also evident in the most recent Intergenerational Report (Frydenberg 2021) which incorporates modelling suggesting that Australia’s migration program not only contributes to per capita GDP growth, but also makes a positive contribution to the budget (because, over the course of their lifetimes in Australia, migrants pay more in taxes than they absorb by way of government services).

Like the PC, the IGR acknowledges that “the contribution of migrants to cultural diversity, community connections and innovation are important but difficult to quantify”. They aren’t counted in any measurement based on per capita GDP.

The IGR also notes that “the economic and social pressures and capacity constraints that result from an inward flow of migrants need to be managed carefully” and that “governments at all levels need to ensure that planning and infrastructure provision keep pace with current and future migration rates”.

That’s something which, I would add, governments at all levels have not always succeeded in doing (and sometimes haven’t even tried to do).

It’s perhaps worth emphasising that these conclusions about the consequences of Australia’s migration program would be rather different if the composition of our migration program had been significantly different from what it has been.

Evidence from cross-country studies suggests that while the impact of immigration on average wages and employment of the “native born” is negligible or positive, it also shows that beneath these averages the impact on different groups can vary significantly.

In particular, the evidence suggests that “selective immigration policies towards loweducated workers will tend to increase inequality between high- and low-educated native workers, while selective policies in favour of high-educated immigration will tend to reduce it” (Edo 2019: 944).

Because of our geographic isolation from most plausible sources of immigrants, we have been able, in John Howard’s notorious phrase, to “determine who comes to Australia and the circumstances in which they do”.

Our immigration intake has been much more highly skewed towards people with knowledge and skills – knowledge and skills which more often than not have been acquired in their countries of origin, at their own expense or at the expense of taxpayers in those countries of origin.

The evidence suggests that Australia’s skilled migration programs – because they bring to this country people who take on roles which complement those performed by workers who are already here, rather than ones which substitute for them – increase both the productivity of ‘native’ workers, and their wages (Crown et al 2020: 597-8).

Such effects are far less evident in the United States or most European economies – where some studies (eg Burzysnki et al 2018) suggest that the more recent waves of immigration have provided smaller welfare gains that previous ones.

This doesn’t amount to a blanket endorsement, on my part, of Australia’s immigration program over the past two decades – or a suggestion that when we do eventually reopen our international borders, that we should simply seek to restore the status quo ante with regard to our immigration program.

In particular, I think the Governor of the Reserve Bank, Philip Lowe, made an important point in his address to the Queensland Branch of the Economic Society last month (Lowe 2021).

This speech has been wrongly, I think, construed by some as a call for a significant reduction in Australia’s migration intake in the post-pandemic world. I don’t think the Governor was saying that at all.

Rather, in my view, he was calling attention to the fact that Australia’s immigration program may not have been as “skills-based” as is widely believed. In particular, he pointed out that a surprisingly large proportion of the 430,000 people who were in Australia on “temporary visas” at the time of the 2016 Census were employed in unskilled jobs including about 18 per cent of all jobs in the food trades, 13 per cent of all jobs in hospitality and in cleaning and laundry, 12 per cent of all food preparation jobs, about 9 per cent of factory process jobs, and 8 per cent of farm jobs.

Dr Lowe made two points about these numbers.

• first, that they probably have contributed to, as he put it, “wages being less sensitive to shifts in demand than was once the case”.
• and second, that the ability to draw upon temporary migrants in these instances “can also dilute the incentive for businesses to train workers to do the required job”.

I think Dr Lowe is probably right on both counts. And as a result, there probably is a case for re-thinking whether we should be issuing as many temporary immigrant visas when our borders re-open as we had been doing previously.

There are some other elements to our migration program which also seem to warrant a re-think. A recent report from the Grattan Institute (which, disclosure, I worked for between 2009 and 2011) (Coates et al 2021) casts significant doubt on the contributions made by the Business Innovation and Investment Program, and the Global Talent Program.

Almost 80,000 people have come to Australia under the Business Innovation and Investment Program over the past decade – perhaps surprisingly, more in 2020-21, a year in which our borders were supposedly “closed”, than in any of the preceding decade.

The Grattan report shows that these people are “older, participate less [in the labour market], have poorer English, and appear to earn smaller incomes than those issued with permanent visas via the points-test or employer sponsorship”. It shows that the program rules which have been purportedly designed to select people with “business acumen” instead result in people buying small businesses with little capacity for innovation.

That may appeal to the quite wide constituency of people who appear to believe that people who run small businesses are inherently more noble, and thus deserve to pay less tax on any given amount of income, than people who work for big businesses, government agencies or other entities. But it doesn’t do anything for the Australian economy.

The Global Talent Program, under which 4,100 supposedly “highly skilled professionals” were granted visas in 2019-20 and another 11,000 in 2020-21 is a particularly opaque part of Australia’s immigration program.

Applicants have to be nominated by an organisation or an individual with a “national reputation” (something for which some industry peak bodies are now charging prospective applicants a fee), or by someone with the impressive-sounding title of “Prime Minister’s Special Envoy for Global Business and Talent Attraction”.

On the surface at least, this appears to be a means of allowing a few people to grant visas to individuals who wouldn’t qualify for a visa under any of the other programs, but who nonetheless know whom to call: which may explain why 75 per cent of the visas granted under this program have gone to people who were already in Australia, and why 75 per cent of applications for them have been processed within 2½ months, whereas 75 per cent of employer-sponsored visa applications take between 5 and 13 months to be processed (Coates et al 2021: 54 and 57).

In the meantime, the closure of Australia’s international borders, in both directions, not just to migrants but to tourists and students – although not, it would seem, to sports people, celebrities (including members of the Kardashian clan) and Cabinet Ministers – is having some other, in my view under-appreciated, economic consequences.

First, it is resulting in a more rapid decline in the measured unemployment rate than had been anticipated by anyone.

Over the decade prior to the onset of the pandemic, Australia’s civilian working age population (that is, people aged 15 and over other than those in the defence forces) increased by an average of 26,000 a month, on average. That meant that, abstracting from changes in the labour force participation rate (the proportion of the working-age population in employment or actively seeking it), the economy needed to “create” an average of about 16,500 new jobs a month for the unemployment rate to decline. As it turned out, the economy created almost exactly 16,500 new jobs a month between the December quarter of 2009 and the December quarter of 2019: so that, although the unemployment rate fluctuated quite a bit over the course of that decade, it essentially ended it the decade where it began, at about 5¼ per cent.

Since the closure of our international borders in March last year, the civilian working age population has grown at an average of 4,300 a month (ABS 2021d). That means that, all else being equal, we’ve only needed to “create” at least 3,000 new jobs a month in order for the unemployment rate to fall.

Of course, in the first few months of that period we lost and then re-created vastly more jobs a month than that. But so far this year – that is, over the six months to June – the civilian working-age population has grown by an average of just under 6,000 a month, implying that the “hurdle rate” of net new job creation in order lower the unemployment rate, given that the participation rate has been essentially unchanged over that period, has been just under 4,000 a month. In fact, the economy has created an average of 47,500 new jobs a month, on net, over the first half of 2021: hence, the unemployment rate has fallen from 6.6 per cent in December last year to 4.9 per cent, the lowest since 2010, in June.

If the working-age population were to continue growing at this rate until June next year, when the most recent Budget assumed that they would then be re-opened, and employment continued to grow at the same rate as it has over the first half of this year, ie 47,500 a month, then all else being equal the unemployment rate would be down to 1.1 per cent by June next year.

That’s pretty unlikely, especially given the lockdowns which have occurred since June and which are now in place – although it’s worth noting that Victoria’s unemployment rate fell in June despite it being in lockdown while the June labour force survey was being conducted, and that not laying off employees is a condition for the receipt of the assistance to businesses which the Commonwealth and State Governments are making available during the current lockdowns. The labour market effects of the recent and current lockdowns are more likely to show up as increased under-employment, an increase in the number of people working “zero hours” but nonetheless counted as “employed”, and a reduction in total hours worked, than as an increase in the “official” unemployment rate.

If the economy were to generate new jobs over the next twelve months at half the rate at which it has over the past six – ie, just under 24,000 a month – then the unemployment rate would, all else being equal, be down to 3.2 per cent by June next year.

That number, too, might be unrealistic. But it surely isn’t beyond the bounds of possibility that the economy could create new jobs at a rate averaging one-quarter of what it did over the past six months – ie, just under 12,000 a month – and if it did, the unemployment rate in June next year would, all else being equal, be down to 4.2 per cent. That’s well into the territory that the Reserve Bank (Lowe 2021: 14) now thinks is consistent with the concept of ‘full employment; and lower than where the Treasury now thinks the ‘non-inflation accelerating rate of unemployment’ or “NAIRU” is (Ruberl et al 2021: 26). It’s also higher than either of them is currently forecasting for that time (4¾ per cent and 5 per cent, respectively).

Of course you may well ask, in the absence of immigrants, where is the demand that will help create those new jobs going to come from?

The answer to that question lies in the second under-appreciated corollary of the closure of our international borders.

The adverse impact of the closure of our borders to in-coming tourists and students on Australia’s tourism and higher education sectors has been widely recognised – and, in the case of the first of these but conspicuously not the second, has been the object of a considerable amount of government assistance. Spending in Australia by foreigners visitors to Australia dropped from $66 billion in the 12 months to March 2020 (just before the drawbridges were pulled up) to $30 billion in the 12 months to March this year, (ABS 2021a) – there apparently being enough foreign students, tourists and others who chose or were allowed to remain in Australia to have spent $30 billion in that period.

What’s been less widely comprehended is the effects of the taboo on Australians travelling overseas (unless they are sports people, celebrities or Cabinet Ministers).

Incidentally, it’s never been obvious to me exactly how banning Australians from travelling abroad helps keep the virus at bay (other than by reducing the demand for quarantine facilities); and I note that to the best of my knowledge, no other democratically-governed country has sought to prevent its citizens from leaving their country as comprehensively or continuously as Australia has done.

In the 12 months to March 2020, Australians spent $56 billion overseas (ABS 2021a). That dropped to just over $1 billion in the 12 months to March 2021.

That $55 billion hasn’t disappeared into the ether.

Instead, it’s been spent within the domestic economy: • on electrical and electronic goods, sales of which were 19 per cent higher in the 12 months to March this year than they were in the previous 12 months;

• on furniture, floor coverings, housewares and the like, sales of which increased by 20 per cent over this period;
• on hardware, building and garden supplies, sales of which were 22% higher in the 12 months to March this year than in the previous 12 months;
• on booze, sales of which were up by 26 per cent in the 12 months to March compared with the 12 months to March last year (this is Australia, after all);
• on clothing and footwear, which were adversely affected by lockdowns for longer than the aforementioned categories of retail, but which in the six months ended March were up by 16 per cent and 13 per cent respectively, by comparison with the six months to March 2020;
• on cars, sales of which were 10 per cent higher in the six months ended March this year than they had been in the same period a year earlier;
• and on renovations, the value of approvals of which by local councils in the six months ended March was 32½ per cent higher than it had been in the six months ended March 2020 (ABS 2021b and c).

Assuming – as seems plausible – that Australians have spent at home in the aforementioned ways or in others (such as “holidaying at home”), all that they would have spent overseas, had they been allowed to, this amounts to a larger stimulus to domestic spending than anything that the federal government has provided through its Budget, apart from JobKeeper.

That’s a big part of where the demand that has helped to sustain more than enough job creation to keep the unemployment rate on an accelerated downward trajectory has been coming from.

And it seems set to continue until whenever it is that Australia’s international borders do re-open.

This sounds almost too good to be true.

And in at least one respect, it is.

There is more than a whiff of “protectionism” about all of this.

Remember that, for the best part of ninety years, Australian Governments of all political persuasions (and the overwhelming majority of the Australian people) thought it was a good idea to force Australians to spend money which they would have preferred to spend on foreign-made goods, instead on goods produced in Australia – even though those Australian-made goods were almost always more expensive, and often of inferior quality, to those made elsewhere – in order, so it was said, to create jobs in Australia, and in particular in manufacturing, it being widely believed that manufacturing jobs were in some way inherently more “worthy” than jobs in other sectors of the economy.

And, indeed, for a time those policies did result in Australia having a larger manufacturing sector than we would have had otherwise.

But, as we eventually came to understand, it also contributed to Australia’s long slide down the ladder of nations ranked by relative standards of living, from (as best as can be determined) near the top at the time of Federation, to 26th or thereabouts in the early 1990s.

After we came to our collective senses and started tearing down our tariff and quota walls, we climbed back to being about 12th in terms of per capita GDP (although of course there were also other factors, not least of them the China-fuelled “resources boom”, which contributed to that outcome).

Now we once again find ourselves forcing Australians to spend at home money that they would have preferred, if they could, to have spent somewhere else. And that is, at least for the time being, helping to create jobs for Australians.

I for one fervently hoped that we don’t become addicted, once again, to a new form of “protectionism”. A misguided obsession with what some like to call “sovereignty” has already led us into some bad policy choices, not least of them the decision to put almost all of our vaccination eggs into one basket because we had the capacity to make it here – a consideration to which the Canadians, in particular, accorded scant regard and as a result of which they are now in a much better position to contemplate the post-Covid world sooner than we are.

Allowing the relatively free movement of people across our international borders, in accordance with criteria that we have determined, has served Australia well for a very long time. There are, I think, some adjustments to the criteria which we applied in the years before the onset of Covid-19 that could be sensibly amended.

But we should avoid throwing the baby out with the bathwater. Australia’s economy, and Australia as a society and as a nation, has been greatly enriched by the contributions that successive generations of immigrants have made.

That doesn’t mean it wouldn’t be sensible to ask whether all of the different components of our immigration program are making positive contributions to our overall “well-being”.

Likewise, it’s counter-productive to pretend that there haven’t been some costs associated with the size of our immigration intake, in particular the effects on housing affordability and on other dimensions of the quality of life, especially in our larger metropolitan areas. That should make us focus on whether there might be other things which we can do to improve housing affordability or reduce traffic congestion (which of course there are – we just don’t want to do them) without having to forego the advantages which a well-structured immigration program can continue to deliver for us.

REFERENCES

Australian Bureau of Statistics (2021a), International Trade in Goods and Services, Australia, May 2021, Canberra, 1 July.

(2021b), Building Approvals, Australia, May 2021, Canberra, 5 July.

(2021c), Retail Trade, Australia, May 2021, Canberra, 5 July.

(2021c), Labour Force, Australia, June 2021, Canberra, 15 July.

Burzysnki, Michael, Frédéric Docquier and Hillel Rapoport (2018), The Changing Structure of Immigration to the OECD: What Welfare Effects on Member Countries?, HCEO Working Paper No. 2018-057, University of Chicago, June.

Coates, Brendan, Henry Sherrell and Will Mackey (2021), Rethinking Permanent Skilled Migration, Grattan Institute, Melbourne, 30 May.

Crown, Daniel, Alessandra Faggian and Jonathan Corcoran (2020), “High skilled immigration and the occupational choices of native workers: the case of Australia”, Oxford Economic Papers, Volume 72, Issue 3, July, pp. 585-605.

Edo, Andrew (2019), “The Impact of Immigration on the Labor Market”, Journal of Economic Surveys, Volume 33, No. 3, July, pp. 922-948.

Frydenberg, Hon. Josh, MP, Treasurer (2021), 2021 Intergenerational Report: Australia over the next 40 years, Canberra, 28 June.

Lowe, Phillip (2021), “The Labour Market and Monetary Policy”, Address to the Economic Society of Australia (Queensland), Reserve Bank of Australia, 8th July.

Productivity Commission (2016), Migrant Intake into Australia, Inquiry Report No. 77, Canberra, 13 April.

Ruberl, Heather, Meika Ball, Larissa Lucas and Thomas Williamson (2021), Estimating the NAIRU in Australia, Working Paper No. 2021-01, Australian Treasury, Canberra, 29 April.