Analysing Australian migration: Getting beyond common sense

I want to give Carrington Clarke, an ABC business journalist, the benefit of the doubt. His piece on Thursday titled “Immigration masking Australian economic decline” hints at an important point often missed but the litany of negative inferences and weak claims which follow, unsupported by any evidence at all, is hard to accept. Clarke highlights how adding more people to the economy is not necessarily good for everyone as individuals. Throughout, the author infers migration to Australia is one reason why individuals are going backwards in economic terms. The way he does this is by using ‘common sense’ arguments about migration, e.g. how more people will push down wages because that’s what Economics 101 dictates.

In the simplest terms possible, if you add an extra person to the economy, it is going to grow in the aggregate. But it may not grow in per capita terms, or for each individual. This is because each extra person may have a positive, negative or nil effect on the average. Say the average wage in an economy is $50,000 but the new worker earns $40,000. She is adding output to the total but reducing the average, ‘dragging down wages’.

Clarke concludes, “High migration makes it nearly impossible for Australia to fall into recession” as migration means the aggregate economy will keep growing.

This is debatable but no-one can deny migration has improved aggregate GDP growth over the past two decades. The net migration trend – the number of people immigrating to Australia minus the number of people who have emigrated – has been high when compared to the 1970s and 1980s. As others have pointed out, net migration to Australia from the Global Financial Crisis onwards may have been the most important factor keeping Australia out of recession alongside the mining boom. From Bob Gregory, an ANU economist and former RBA board member:

“Our extra ordinary economic success since the GFC owes a great deal to the increased level of national income produced by the unforeseen population expansions generated by our new immigration program.”

“It is possible that the economic magnitude of the immigrant policy change over the last decade has been as large as the mining boom impact.”

(Source: A good Chris Bowen speech from September 2016)

This is essentially Clarke’s point but while he is focusing on the potential negative effect on the individual, others focus on the positive national effects. He takes this point to an extreme by saying it is ‘nearly impossible’ for Australia to experience a recession. Given the recent aggregate GDP figures, and the declining rate of net migration, I don’t buy this claim as migration trends can change relatively quickly and it is entirely possible Australia could fall into recession.

And this is my main concern. By failing to explain why migration has changed, Clarke isn’t able to place these migration trends in context and explore the question. If you can’t do this at the ABC, where can it occur in mainstream journalism? You shouldn’t be able to discuss immigration and economic changes since the 1980s without noting the transformational policy changes that have occurred. Migration changed from a process controlled strictly by governments to a mixed system where economic demand itself influences the rate of migration as well as government decisions.

Regardless of an interesting main point, this is no excuse for poor supporting analysis without any substantial evidence.

Claim One: “The truth is, migration to Australia is still proceeding at a record clip.”


(Source: ABS, Net Migration)

Net migration figures show quite clearly immigration trends since the high of 2009 are slowing. This is akin to the rate of GDP growth slowing but the economy as a whole still growing. When the latest GDP figures come out, there is almost universal focus on the rate of growth as opposed to the size of the economy. This is sensible as in many cases the total size is less relevant than the rate of change. In migration, the equivalent is total number of migrants in Australia, or the stock figures of various visa categories.

Clarke uses the stock of temporary migrants in Australia as evidence. While the table he highlights notes 12 month changes, he fails to mention permanent migrants (accounting for about 40 of net migration) as well as any emigration at all (where citizens and others leave Australia). Net migration, not the total number of temporary migrants, is the primary tool used by policy makers to assess migration trends. The latest estimates of net migration from the ABS in December 2016:

“The preliminary estimate of net overseas migration (NOM) for the year ended 30 June 2016 (182,200 people) was 3.0%, or 5,300 people higher than the net overseas migration recorded for the year ended 30 June 2015 (176,900 people).”

Is 180,000 a record clip? That’s about 0.75 per cent of the population on an annual basis. The latest Intergenerational Report helpfully puts this into a historical context:


A net migration rate of 0.75 per cent is lower than the last decade average of 1.1 per cent as well as the post-war net migration rate of 1.0. It falls somewhere between the heights of post-war Australia and the lows of the 1970s to 1990s.

All of this means the “truth” is migration to Australia is not proceeding at a “record clip”. In fact, it is falling in trend terms and perhaps people should be talking about this more as it likely reflects a soft labour market and economy. This is particularly true if you exclude international students (which are growing strongly) and focus on visa categories which are influenced by relatively global strength and domestic labour demand, like 457 visas, working holiday makers and New Zealand citizens.

Claim Two: “New workers mean greater competition for jobs, which suppresses wages.”

Unfortunately Clarke doesn’t provide any evidence for this claim, except to note how wage growth has been falling in recent times. It’s true migration could be affecting wage growth but all the evidence we have suggests this is not the case.

There is a pretty clear consensus among labour economists that migration – regardless of the workers who are joining an economy – doesn’t have large employment or wage effects, on average. What matters is composition: “the skills of migrants, the skills of existing workers, and the characteristics of the host economy” (source: Martin Ruhs, Migration Observatory, United Kingdom). In Australia, this is the rationale behind a skilled migration program. If we take Clarke’s point literally, more doctors and IT specialists will reduce the wages of doctors and IT specialists, making it cheaper for people to see a doctor and for companies to get IT assistance. Then, these new doctors and IT specialists will spend their incomes on goods and services, rising the demand for things like hairdressers, accountants and construction workers. When you throw all this in together, any number of scenarios are possible, including positive, negative and no effects.

The real take away is to consider demand and supply effects, with regard to the labour market fundamentals which already exist and the characteristics of new migrants. While I acknowledge there isn’t much space in an 800 word analysis piece, it’s not hard to note some of these trends and effects in passing.

So what does the evidence say? Here are a couple of statements which show why Clarke’s claim about suppressing wages should be taken with a large piece of salt:

Productivity Commission (2016):

“International studies find that the overall impact on the local labour force is small (either positive or negative).”

“Research commissioned for this inquiry suggests that over the decade since 2001 (generally a period of robust economic growth), on balance and in aggregate, recent immigration had negligible effects on the labour market outcomes of the local labour force.”

Migration Observatory (2016):

“Empirical research on the labour market effects of immigration in the UK suggests that immigration has relatively small effects on average wages but more significant effects along the wage distribution, i.e. on low, medium and high paid workers.”

Claim Three: “More people also mean more demand for scarce goods and services. When there’s already a tight supply of a particular good, it can mean huge price rises.” (talking about housing)

It’s undoubtedly true that demand for scarce goods and services leads to higher prices. But as has been acknowledged by almost every serious policy guru in the country, the reasons for housing scarcity are numerous, including demand-side policies like negative gearing and capital gains tax, as well as supply pressures such as zoning and urban planning. Lumping Australia’s housing market onto new migrants is a particularly low blow when this was a major election issue.

Yes, more people will increase demand for housing and other services. But as the graphs above show, migration has not come out of nowhere and is a known quantity. Planning for this, instead of passively ignoring it, is a much better approach than scapegoating migrants. Understanding migration flows as response to underlying economic conditions, instead of an exogenous force or shock imposed by governments, helps understand why blaming migrants is not the way forward on housing policy.

Claim Four: Migration ‘Good for business, governments’ but “But it isn’t necessarily good for ordinary workers.”

As with the effects on wages and employment, this question is compositional. It depends where you live, what you do and what your skills are. However this is the very reason underpinning Australia’s skilled migration program. It is deliberately set in the “national interest” or for ‘ordinary workers’. Skilled workers increase supply at the top of the labour market, most assisting those at the bottom of the labour market. Skilled migration is a mitigating force on inequality in Australia, albeit likely a small one (but opening up more low-skilled opportunities to Australia would better assist global inequality instead of national inequality). For example, while 457 visa holders make up less than one per cent of the labour market, their salaries are about 20 per cent higher than the average full time wage. English-speaking migrants do better on average than the average labour market participant while non-English speaking migrants do slightly worse. In addition, the economic demand created by new migrants is propping up entire industries – construction and higher education come to mind – and while this is not a reason to run a migration program, it must be considered when assessing the effect on ‘ordinary workers’. Clarke also fails to note how migration may drastically improve the lives of new migrants themselves. Should we consider this? Again, it depends on the policy goal of governments.

The real people who feel the biggest brunt of new migrants are likely the migrants who are already here. Numerous studies show this and it’s because these groups are often the most similar in any economy, therefore having higher rates of substitution on employment and wages.

Normally, any analysis of non-asylum migration to Australia is a positive as it is drastically underexposed in mainstream media outlets. Kudos to Clarke for taking the time to write about an interesting aspect of migration to Australia, the differences between aggregate and marginal benefits/costs of migration. However it would have been good to see a broader discussion about why migration is the way it is, a broader set of possible effects and some of the well-established evidence about these questions. Particularly for analysis on the ABC, which should be held to a higher standard, the four claims above deserve more scrutiny than they received.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s