Speech to the Migration Institute Australia: Economic impact of migration

Today I gave a speech to the annual Migration Institute of Australia conference. Below are my prepared notes however I did deviate a little bit when speaking. I used to hate public speaking but now I really enjoy it, especially with an engaged audience who are knowledgeable on the topic of discussion.

Thank you to the MIA for inviting me to speak. My name is Henry Sherrell, I work for the Migration Council Australia. Before this I worked at the Department of Immigration and Citizenship.

After the population ‘debate’ that occurred around the 2010 election, there was a significant chance the next Intergenerational Report would kick off a game of ‘blame the migrant’ where migration becomes the public scapegoat for every single policy failure across multiple government jurisdictions: jobs, congestion, housing, healthcare, taxation. You name it.

The implication behind this public sentiment is that migrants are bad for Australia. This was the reason we decided to write a report about the economic impact of migration, bringing out of the shadows some of the effects that many people take for granted.

While many in this room will be well aware how the general migration program is biased towards skills and has a positive economic effect, the notion that migrants steal jobs and bludge off social security remains potent in some sections of society.

In a recent ANUPoll this year, 29 per cent of the population agreed migrants took jobs that should go to Australian born people (a majority disagreed with this statement). Further, there are those who will seek to exploit this in periods of economic uncertainty, something more likely than not in the coming years.

The last major attempt to canvass the economic impact of migration was undertaken by the Productivity Commission in 2006. However this report as is understandable, drew on assumptions that were derived of the era – lower trend numbers and lower average skills. A decade is a long time in public policy and these trends have changed. There was a particular emphasise on permanent migration.

However before I get to our report, it is worthwhile providing a bit of context to questions on the economics of migration.

The study of the economic effects of migration has grown quickly in the last two decades. David Card kick started the trend with his study into the Mariel Boatlift.

This is a seminal event for migration analysis. This is when 125,000 Cubans left within 6 months for the US. More than half of these people settled into Miami. Looking at Census records, Card found that a 7% increase in the low skilled labour market in 6 months for Miami did not have a negative impact on existing workers. While there were changes for individual workers, on average, both wages and unemployment remained unchanged (note the equivalent figure for Sydney would be somewhere in the range of 200,000 people in a 6 month period, more than the entire net migration rate for all of Australia over 12 months).

The research has become more sophisticated in the 20 years plus years since Card published his paper. If you are not a labour economist – and I’m not – it’s nearly impossible to read these papers. But the findings have remained largely the same, regardless of what country you look at or what labour market or what type of migrants.

Common migration trends do not result in negative effects for existing workers.

Underlying so much of the research is the concept that there isn’t simply a fixed number of jobs in an economy. Migration is not zero-sum and migration can be complementary as opposed to substitutes. A good analogy is the large increases in female labour force participation over three decades. Women did not take existing jobs that males previously worked in. In a dynamic, flexible economy, things sort themselves out over time.

If you take the migration trends from 2000 to 2013 and average them out, migration to Australia between now and 2050 will account for:

  • 7 per cent to participation rate
  • 9 per cent per capita GDP growth
  • 14 million people

On average, ten migrants will have the same economic effect of eleven existing residents.

These findings are based on modelling from Independent Economics and are the difference between what would happen with current migration trends and what would happen with basically a zero net migration rate.

I want to stress it is unlikely we will see a net zero net migration rate. However the purpose of the research was to account for the effect of migration as a whole and this comparison allows for that most effectively.

I’m going to get a little bit into the details because I think it’s important to the debate. There are a couple of key assumptions. The first is the rate of net migration and the second is the composition of the migrants.

We based the net migration rate on the trend between 2000 and 2013. It’s quite difficult to forecast net migration. Some might say it’s impossible. Our estimate contrasts to those used by the Treasury for the IGR. The IGR forecast a flat figure of 215,000 over 40 years. This figure starts off as about one per cent of our population at the start and gradually falls to 0.6 per cent by the end of the forecast.

This doesn’t look right to us. Why would net migration fall by almost half over a 40 year period? Instead, we used a forecast of 250,000 initially and transitioned to 0.85 per cent of population each year from 2030.

Of course, trends do change. But we know one thing for certain. Without a migration program, the population stagnate at about 24 million.

The model used is comprehensive. Attributes include; Labour and skills, participation, terms of trade, investment, financial wealth, natural resources, infrastructure, education productivity, R&D productivity, government expenditure.


Population in general gets a bad rap. We talk about overpopulation and sustainability almost came to mean ‘no new people’ a few years ago.

A quick glance of countries where the population is stagnating highlight the potential economic dangers. Italy and Japan are the two most common examples in this. We will soon add China to this list.

Australia is not amongst these countries. Our report projects a population of 38m in 2050, rising to 40.1m by 2055.

This does not sound like the Australia your parents grew up in. Yet it is important to note this is based on current trends. This is happening now. Infrastructure and urban planning are going to become more important, not less, in this future.

For too long, migration has been set aside by key policy areas. Instead of calling for sustainable population levels, integrating migration policy and population policy better into key priorities for state and federal governments will improve the transition for a growing Australia.


Labour participation is projected to be 15.7 per cent higher with migration than without in 2050. This means current migration trends are worth about 0.4 percentage points of labour participation each year.

This is primarily driven by two factors:

  • Migrants are younger than the average worker.
  • Migrants are more skilled than the average worker.

Again, this is based on trends and perhaps things will be very different in 2050. But on current policy, labour participation gets a big boost from migration. More directly than a population figure, participation points to how well the labour market will respond to an older society.

These effects offset some of the marginal costs of an ageing society. Migration cannot ‘fix’ or ‘solve’ an ageing population. But it can slow the rate of change and afford policy makers critical time to make adjustments.

These modelling results build on what we already know. A short paper from Mark Cully on the decade from 2000-10 showed similar findings. Increases in labour participation for that period were driven, at least in part, by the effects of policy change by previous governments.


The effect of migration on productivity is complex. While we project an overall increase in GDP per capita of 5.9 per cent by 2050, adding more people to a labour market hurts labour productivity if it is done in isolation.

This is because an economy takes time to adjust to change. Migration is no different. The model found a labour productivity decrease of 7.9 per cent because of the time it takes for capital to adjust to a larger population.

This points to a complex picture for productivity and a role for policy makers. If current trends bear out, reducing the barriers to capital adjustment for a growing labour market will enhance productivity. As a small, open economy, this is one of the most important complementary measures when running an open migration program. Speeding up the capital adjustment process will smooth the transition process as the labour market grows and underpin labour productivity.


The “Three Ps” tend to dominate these discussions. But we thought it was also important to have a look at the distributional role of migration given the increasing focus on inequality.

The distributional effects predominantly come back to a previous point. Migrants in general are more skilled than the average worker.

A skilled migration program adds labour disproportionately to the skilled labour market. This additional supply creates slightly more competition amongst skilled workers while also adding more demand for lower skilled workers. Overall this effect reduces income inequality at the macro-level.

Previous studies show the effect of migration in Australia on wage growth for low skilled workers could be as high as double the effect on high skilled workers. Our projections show a similar finding. This is a critical point often missed in the public debate.

Wrap up projections

To recap the economic effects to 2050 based on current trends against a hypothetical situation of no migration; 14 million extra people, a 15 per cent difference in labour participation, 5.9 per cent GDP per capita difference and a helpful method to mitigate income inequality.

The future

So what about the future and how might this economic contribution be enhanced? I have two quick suggestions I’d like to put forward that can better capture some of these potential gains. One relates to migration globally and the other here in Australia.

The world is becoming more mobile. Migrants make up on average about 10 per cent of OECD population. This is more than double since the end of the Cold War.

Australia is already ahead in many measures. Our diversity and social cohesion mean we can attract a unique blend of migrants. But policy makers cannot rest as global competition to attract migrants will only increase in the future.

While some advocate for a ‘world migration organisation’ akin to the WTO, I don’t think this is a viable idea for a range of reasons. But Australia can and should be doing more to formalise regional and bilateral migration links. Treaties have traditionally been the domain of security and trade policy. Yet to get the most out of future migration flows, formalised migration links will be necessary.

A great example exists, hidden in plain sight. A recent study from the ANU into the economic effects of the United States Australia Free Trade Agreement found there was little net benefit. However this ignored one critical piece of the USAFTA. As American’s already have excellent labour mobility to Australia through the 457 visa program, Australia was able to negotiate the “E3” visa. While the rest of the world has a total of 85,000 H1B visas to enter into the U.S., Australia has 20,000 to itself.

I was speaking to an Australian who works at Instagram in Washington D.C. a couple of weeks back. Their business is growing very quickly but visa restrictions means London has put on 1000 staff that they would prefer in California. As an Australian, he had no trouble with his visa. Thinking more thoughtfully about these issues will mean Australia can better integrate into a growing global marketplace.

Closer to home, there is more we can do to support new migrants to increase their contribution. Helping people to help themselves, and the economy. This is predominantly the domain of settlement support.

We know English language is the most important variable in how well an average migrant succeeds economically in Australia. With increasingly diverse flows of migrants, we need to think carefully about how this is done.

For example, the spouses of 457 visa holders are not eligible for English language support. Traditionally this is because the 457 visa is temporary in nature. But increasingly, this is an administrative description at odds with the labour market and society.

A 2012 survey of the 457 visa program showed nearly four in five non-English speaking migrants intended to stay in Australia. Spouses of these migrants are likely to become permanent migrants and future citizens.

It is common knowledge the best time for settlement support is right at the start.

If we are going to promote the ‘two-step’ migration model – coming to Australia on a temporary visa and then moving to a permanent visas – we need to also carefully assess where additional support might be required and how to best provide.

As I end, I want to acknowledge that the economic effects are just one part of Australia’s migration story. The social and cultural – not to mention humanitarian – motivations are equally strong. It is up to policy makers and governments to determine the priorities. Perhaps we have moved too far to emphasising economic narratives? Perhaps we haven’t moved far enough? These are questions that will play out over the following decades, particularly questions around low skilled migration and migration for high end services, such as health and education.

Thank you.

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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