How to feed research into Australian migration policy: Self-selection and labour market performance

I attended a great presentation yesterday by Christian Dustmann, who heads up the migration-themed CREAM research organisation, based at University College London. I’ll post a link to the video when available. There were lots of interesting bits and pieces but two things stuck out in an Australian policy context.

The first point was his emphasis on how migrants self-select and the flow on effects on these processes. There is lots of research on migrant selection but I find it tends to fall into the background when thinking about migration policy responses. The act of migration itself is a choice. Even in forced migration situations there are often options about where and when to travel. Dustmann focused on the selection at play in labour markets. Migrants are not stupid he said, and will select into markets where outcomes are more promising.

In Australia, this has policy implications. For example governments of both stripes and others on the crossbench continue to promote migration as a partial solution to regional labour shortages and even propping up the viability of existing communities. But it is hard to push migrants out into regional areas as there are relatively fewer employment opportunities. Migrants in Australia tend to be more urban than even existing Australian residents, selecting into locations which have potentially greater incomes. Recent policy examples such as the backpacker tax and the SHEV protection visas shed some light on this. Backpackers wouldn’t be out in regional areas if it weren’t for government policy. The incomes are not great and the work is hard. They select in looking for more residency. Instead of using prices and wages, the government can use the promise of more time in Australia. This obviously doesn’t work with Australian workers as they already have unlimited residency. The same forces are at play with the SHEV visas. The government knows asylum seekers are searching for more permanent residency outcomes so load up regional incentives into a more promising visa pathway.

Yet there are clear limits to these policy incentives. It is much more difficult to create this same incentive for a highly skilled worker who has numerous options across the world as they will select in elsewhere. Relying on this incentive is also quite short-term as after permanent residency is granted and labour mobility frees up, you would expect a drift away from these labour markets in line with choices others would make. Using migration to tie people exclusively to a single region is also not a good idea over a long period. Dustmann’s point was this process is more complex than people and policy-makers probably imagine. Instead of just focusing on the effect on wages and ‘Australian jobs’, existing native residents might leave labour markets or choose not to participate when migrants enter. This is not a simple static environment but one where change and response to that change is the default. Unfortunately this more complex argument is not very compelling to people who believe migrants steal domestic jobs from Australian workers. The response from both the Coalition Government and Labor must improve to counter a broader acceptance that migrants steal jobs, as the media certain isn’t going to help. And as the backpacker tax argument shows, often some of the more disadvantaged socio-economic communities have come to rely heavily on migration despite increasingly voting for political rhetoric which says there is no place for them in modern Australia.

In a roundabout way, this leads to the second take away.

Australia runs a skilled migration program and the general consensus is we do it well. The visas themselves are mostly achieved by using proxies for determining skills, things like a qualification or demonstrated work experience. But using proxies does not guarantee migrants will work in jobs associated with those qualifications or earlier work experience. This isn’t necessarily a bad outcome but we know migrants face added barriers in the labour market not faced by native workers, things like language proficiency and understanding norms. This leads to ‘downgrading’ in terms of their jobs which is difficult to capture in research and data because we never have a full picture of how people fare.

Dustmann presented some of his recent research using earned income and compared this to native income in a distributional sense. The graph below shows the proportion of recent migrants at different points in the earned income distribution, compared to natives.

dustmannwagespredictionSource: Dustmann et al, The effect of immigration along the distribution of wages

The flat line represents natives while the two other lines represent the actual and predicted outcomes of migrants place in the wage distribution of the United Kingdom.

The dotted line (actual) shows recent migrants are 1.5 times more likely than natives to be working at the very bottom of the wage distribution, which then falls away, flattens out and then rises at the very top end of the distribution. This pattern is quite different to the other line, which is the predicted distribution of recent migrants using their education as a proxy. Effectively, for recent migrants arriving in the UK, this graph captures the actual downgrading which occurs, at least for income purposes.

This is fascinating for Australian policy makers. Running a skilled program and relying on education and previous experience requires bureaucrats to make a whole bunch of decisions, such as what occupations to put on eligibility lists for example. You might have heard Peter Dutton recently say goat farmers are now not allowed to be hired on 457 visas. This is a policy choice. What this data shows is downgrading has massive implications for policy makers because out in the labour market, employers and migrants sort themselves out differently to what you would predict if you just considered qualifications. Thinking about this type of finding and replicating it for Australia should help us better understand migrants in the Australian labour market. If anyone reading this is interested in looking into possible replication of this research in Australia in 2017, please get in touch ( I am following up whether the data is available and what might be required. The paper and methodology are linked above in the source. It doesn’t look easy but I think it is worth considering given the potential implications.

A big concern I have is how research about migration is getting fed into policy initiatives. I think we tend to do this fairly poorly and while migration policy cannot address everything which happens in the labour market after a migrant arrives and starts working, constantly trying to work out how to improve the situation is important. I don’t see much evidence this is occurring in any structured and meaningful way, despite the large amount of hot air which being spewed forth in the wake of Trump and the 2016 Australian election. Without it, simple reactions – like Dick Smith’s recent claims – to complex environments will start driving political decisions, leading to disastrous unintended consequences.


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 )

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s