skilled migrants, skilled migration, Australia, permanent residence, test score, state selection, selection policies, migrants to Australia, market absorption, Labor market, applicants, visa entry, skilled job, migrant, Immigrants to Australia, Visa Category, hairdressers, skilled migration visa, reference group, migration programs, migration flows, government, The government, MODL, national government, economic migration, migration policies, Australian Department of Immigration and Citizenship, temporary residence, Australian Department of Immigration, Skilled migration selection policies, skilled occupation, OECD countries
Skilled migration selection policies: recent Australian reforms Mark Cully1 Introduction Australia has been operating migration programs that are selective on the basis of skill since the 1970s. They are selective in the sense that certain threshold requirements must be met to be eligible for skilled migration for example, an applicant must be recognized as competent to meet the Australian work standard for defined occupations and also in the sense that applicants can be ranked and prioritized, which then determines if and when a skilled visa is granted. The policies that determine the selection process have evolved over time. While they are complicated in their detail, at heart they favor those who are young, more qualified and experienced, more fluent in English and with skills in demand among employers. Many countries now look to Australia, and other traditional settlement countries such as Canada, to emulate such policies. This article
investigates whether skilled migration selection policies work, using Australia as a case study
. It also provides an overview of recent reforms. A brief outline of skilled migration in Australia Since white settlement began in 1788, Australia's population has been replenished with wave after wave of migrants. The lowest ebb was in the years following the Great Depression
leading up to World War II. At the end of the war, migrants made up 10 per cent of the population. The Australian Department of Immigration was established in 1945 and since then more than seven million people have been granted permanent residence. Behind Luxembourg, Australia has the second highest density of overseas-born in its population among OECD countries
, around 27 per cent. Skilled migration to Australia can be either permanent or temporary, and can be characterized as a "hybrid system" (Papademetriou et al. 2008). Traditionally, applicants for skilled migration were selected on the basis of their attributes and capabilities; permanent residence was granted with no requirement to have arranged an offer of employment 1 Mark Cully is Chief Economist at the Australian Department of Immigration and Citizenship. This article draws heavily on Cully et al. (2011), which is joint work with Kee Lim, David Smith and Caroline Levantis. The views presented in this article are those of the author and are not to be attributed to the Australian Department of Immigration and Citizenship or the Australian Government.
beforehand. Selection under this route is by application, assessed using a governmentadministered points test. For shorthand, we call this "supply-driven" migration. Since the mid-1990s, policies have altered to embrace "demand-driven" migration hence, the hybrid. This was done in two ways. First, by giving greater weight in the points test to applicants whose skills were in demand among employers. Second, employers have been given, subject to certain eligibility conditions, the ability to themselves select migrants through employer sponsorship, for either permanent or temporary residence. Temporary skilled migration is entirely demand-driven. Employers who are unable to fill a skilled vacancy within their local labor market
can sponsor a migrant, subject to meeting sponsorship obligations and paying the going rate for the job. There is no cap on the number of visas that can be granted. Many temporary skilled migrants go on to become permanent residents through one of the several possible pathways open to them, most typically through their employer sponsoring them for permanent residence. In 2010-11, the number of skilled migrants coming through the demand-driven route, whether permanent or temporary, was almost twice as large as that coming through the supply-driven route, 66 900 compared with 34 900. The cumulative operation of skilled migration programs has transformed the character of the Australian workforce. At the time of the 2006 population census, migrants made up more than a quarter of the working age population, namely those aged 15-64. Among these migrants, almost three in ten held degrees, compared with less than one in five Australianborn. Migrant selection The process by which skilled migrants are selected can be separated into two discrete decision-making stages. In the first stage, the would-be migrant must determine that they wish to leave their home country for another country. In the second stage, a destination country must determine to accept the migrant. The first stage (self-selection) is important because would-be migrants will have different attributes to those who have no wish to migrate. For example, they will differ with respect to age and skill, and perhaps also across attributes such as pluck, tenacity and love of adventure that may be predictive of success. They will most likely also differ to those who are willing to migrate but to a different country. In pioneering work, Borjas (1987) developed a model to characterize the self-selection of migrants as either positively or negatively selected. The positively selected are those who come from the upper end of the skill distribution in their home country. They do so because the pay off to their skill is likely to be greater in the destination country. The negatively selected are those at the other end of the skill distribution who migrate because the low skill penalty is less punitive in the destination country. For example, a relatively high minimum wage would encourage negative selection. The implication of Borjas's model is thatmigration flows
are driven by a range of factors that may be quite independent of migration policies.
It is only at the second stage (state selection) that a national government gets to impose its own migration selection policies. The effectiveness of state selection can only be judged against the counterfactual of what kind of migrants would have been chosen, and their associated settlement experiences, if selection at the second stage had been entirely random. For example, as shown above, educational attainment has risen among successive waves of immigrants to Australia, but we do not know how much of this is attributable to selection policies placing greater weight on skill and how much to an increase in the number of skilled people wishing to migrate to Australia. There are many empirical studies that investigate these issues.2 For the most part, they find the attributes of migrants entering under skilled programs differs systematically from those entering under family reunification programs. They also mostly find that skilled migrants perform better in the labor market, and that these differences largely disappear once the analysis accounts for age, qualifications and language proficiency. There is succor here for both academics and policy makers. For academics, visa category seems to have no or little bearing on the returns to education, while for policy makers these studies show
that countries can alter the composition of their migrant intake by preferencing those that it favors and screening out those it does not.
The assessment of the preceding paragraph is not universally held. Jasso and Rosenzweig (2009) compare Australia and the United States and find "no evidence that the differences in the selection mechanism used to screen employment migrants in the two countries play a significant role in affecting the characteristics of skill migration." Even if this finding is true, differences could arise if, as they do, the two countries assign different fractions of available migration places to economic migration versus family re-unification. Recent reforms to skilled migration in Australia
From 2009 onwards the Australian government embarked on a series of reforms to skilled migration. Using the framework outlined in the previous section, these reforms can broadly be characterized as screening for negative selection and sorting for positive selection. Rather than catalogue the full range of reforms, in what follows two are described in some detail.
As discussed earlier, the hybrid of supply-driven and demand-driven skilled migration was first given expression in the points test through assigning bonus points to applicants whose nominated occupation featured on the Migration Occupations in Demand List (MODL).3 The list was introduced in 1999 to make the skilled migration program more directly responsive to labor market needs. It was updated twice yearly on the basis of evidence gathered from surveys of employers advertising skilled vacancies.
For Australia, see Antecol et al. (2003), Cobb-Clark (2000), and Miller (1999).
All primary applicants for skilled migration need to nominate a skilled occupation and be pre-assessed as
having the necessary competence to perform that occupation to the Australian standard.
The government used the opportunity provided by the global economic crisis to institute a review of the MODL, prompted by concerns that it was no longer meeting its purpose. Among these concerns was the blow-out in the number of occupations on the list, which had reached over 100 by the end of 2008. It was also the case that would-be migrants and educational institutions
had realized there was an almost seamless pathway for international students to attain permanent residence if they enrolled in a course of study which would qualify them for an occupation featuring on the MODL. The review of the MODL found that it was an inefficient tool for assisting employers to remedy skill shortages. This was because the lag time between a new skill shortage becoming manifest, evidence of this becoming sufficiently compelling that the occupation was added to the list, applicants incorporating this into their decision-making then migrating, probably took two years to unfold. In contrast, take-up of the temporary skilled migration visa had grown rapidly and it had demonstrably proven its worth in quickly enabling employers to fill job vacancies where no locals could be found. Rather than persist with using the points test to give expression to the hybrid character of skilled migration to Australia, the government decided to cleanly divide the supply-driven route from the demand-driven. The MODL was revoked in February 2010 and along with it the bonus points in the points test. From this time, demand-driven skilled migration was to be met through the temporary skilled migration visa and the permanent employersponsored visa. If that was to be the case, then what role was there for supply-driven migration? Here, the government decided to position supply-driven migration within its overall workforce development strategy. A newly established agency, Skills Australia
, had been tasked with providing the government with advice on public provision of post-school education and training
. It took the view that government ought to intervene only in the area of specialized skills, leaving it to the market to resolve other skill shortages. Specialized skills were defined as those that: took several years to learn; had high correspondence between a field of study and employment in a given occupation; resulted in high economic and/or social costs to local communities
if the skill was in short supply;4 and, had reliable information on which to make the preceding judgments. Within this framework, Skills Australia was asked by the government to advise which occupations met these criteria and supply could be sourced, in part, through migration. Some occupations were deemed ineligible for migration, for example because the work needed to be undertaken by an Australian citizen. Others were deemed temporarily ineligible because of evidence of an ongoing over-supply, the principle being that migrants should not displace Australian workers. The number of occupations for which skilled migrants could nominate was cut from more than 400 to around 180. Fashioned this way,
The archetypal example here is hairdressers. It may take several years to become a proficient hairdresser,
it may be the case that most people studying hairdressing become employed as hairdressers, but: are
there significant social and/or economic costs
to communities if hairdressers are in short supply?
the role of supply-driven migration is to complement the provision of post-school education and training to ensure an adequate future stream of specialized skills. Would-be migrants coming through this route were still subject to the points test. A review of the points test was conducted and a revised test took effect from July 2011. The government also announced that it intended to fundamentally reform the administration of the points test, by prioritizing applicants according to their score. Up until now, applicants who meet the pass mark are granted a visa. In recent years there were more applicants than places available for skilled migration, resulting in a queue. The new system dispenses with the queue. In practice what this means is that applicants will be sorted on the basis of their points test score. To put this scheme into effect, a new two-phase processing system is being introduced from July 2012. In the first phase, applicants will submit electronically an expression of interest in skilled migration, providing sufficient information from which to derive a points test score. In the second phase, people will be invited to apply for a skilled migration visa in descending order of their points test score. In effect, the points test pass mark from one year to the next will serve as a kind of equilibrium price, with the volume of invited applications roughly balancing the volume of these skilled visas allocated by the government on an annual basis. Labor market integration
of recent skilled migrants Informing the development of these policy changes was evidence from a recurrent survey of recent migrants to Australia. Since 2009, cohorts of recent migrants have been surveyed twice a year. The focus of the survey is on the labor market absorption of new migrants. Cully et al. (2011) pool data from three cohorts to investigate labor market absorption for different categories of new migrants. We follow an approach used by Aydemir (2010) in analyzing short-term outcomes for new Canadian migrants. Our interest is in testing the efficacy of policy settings. What we therefore wish to estimate is, independent of human capital
characteristics, the "effect" of the visa entry category on labor market absorption. Of course the outcomes are not independent of human capital characteristics; however, it is these precise characteristics which are determinative in the state selection stage of skilled migration. Table 1 reports some results from this work. It shows the marginal effects of the probability of being employed in full-time skilled work for different visa entry categories, and earnings differentials, relative to a Reference Group
of migrants entering as partners under the family reunification program. The rationale for using these migrants as the reference group is that they essentially go only through the self-selection stage. So long as the Australian authorities are assured of the bona fides of their relationship, and other necessary checks are satisfied, a visa is granted. If state selection was having no effect then differences in labour market absorption ought to be minor, recognizing of course that some partners will be more interested in establishing their home in Australia than immediately seeking out work.
Table 1: Estimated labor market outcomes of recent migrants
Full-time skilled Weekly earnings Full-time skilled Weekly earnings
Source: Cully et al. (2011) Notes Reported as marginal effects. All results are statistically significant at 0.01. Reference group are partners in the family stream. Results are for migrants aged 18 to 45, and control for: whether a former international student; date of survey; years in Australia; state of residence; marital and dependents status; and, whether born in a mainly English speaking country or otherwise. The earnings equation is estimated for employed migrants only, and includes the same controls plus whether the migrant is employed full-time.
As can be seen from the table the effects are very far from minor. For example, a female migrant coming through the employer sponsored route was 74 per cent more likely to be employed full-time in a skilled job than an otherwise comparable female migrant coming through as a partner in the family stream. If both were employed, the earnings differential was 52 per cent. Across the three skilled entry categories examined here skilled independent and state sponsored are sub-sets of the permanent supply-driven route all had much superior outcomes to otherwise comparable partners, with the best outcomes enjoyed by those who were employer-sponsored.5 This provides support for the shift more than a decade ago towards embracing demand-driven skilled migration.
To reiterate again, the "otherwise comparable" construct adopted here ignores differences in education and experience for the very reason that these feature in the decision to grant a skilled visa. If these characteristics were also controlled for, much of the difference in outcomes shown in the table would evaporate. That issue the returns to education and experience that accrues to migrants is an important issue, but it is a second-order issue for policy makers relative to the issue of whether state selection is effective in promoting positive selectivity.
Cully et al. (2011) use the same survey data to examine the impact of the new points test. A new points test score is derived for survey respondents
who had passed the former version of the points test. The new points test score is then used as an explanatory variable in an earnings equation. From these results it is possible to predict the average earnings of skilled migrants at different points test threshold values. The results show earnings are positively associated with higher threshold values. This provides qualified support for the new selection system which will take effect in July 2012 of choosing first those applicants with
The results found here are stronger than those for Canada reported by Aydemir. Some of that could be
attributed to the Canadian data capturing migrant outcomes at up to two years after entry, as it is well
established that differences tend to narrow over time.
the highest score, noting that the results are biased as the survey excludes, by definition, those who might pass the new points test but did not pass the former test. More definitive evidence will become available once the new arrangements are in place. This article has shown that for a country like Australia, which is blessed in having more people wishing to migrate to it than the places it makes available, migrant selection policies do work: they deliver markedly superior labor market outcomes than would accrue if wouldbe migrants were chosen at random.
Antecol H., Cobb-Clark D. and Trejo S
Immigration Policy and the Skills of Immigrants to Australia, Canada and the
United States. Journal of human resource
s 38 (1), 192-218
Immigrant Selection and Short-Term Labour Market Outcomes by Visa
Category. IZA Discussion Paper No. 4966
Self-Selection and the Earnings of Immigrants. American Economic Review,
77 ( 4), 531-553
Do Selection Criteria Make a Difference? Visa Category and the Labour
Market Status of Immigrants to Australia. Economic Record, 76, 15-31.
Cully M., Lim K., Smith D. and Levantis C.
Do Skilled Migration Selection Policies Work? A Case Study of Australia.
Department of Immigration and Citizenship. Staff Research Paper
Jasso G. and Rosenzweig M.
Selection Criteria and the Skill Composition of Immigrants: A Comparative
Analysis of Australian and US Employment Immigration. In: Skilled Migration
Today: Phenomenon, Prospects, Problems, Policies (J. Bhagwati and G.
Immigration Policy and Immigrant Quality: the Australian Points System.
American Economic Review, 89 (2), 192-197
Papademetriou D., Sommerville W. and Tanaka H.
Hybrid Immigrant-Selection Systems: the Next Generation of Economic
Migration Schemes. In: Talent, Competitiveness and Migration, Transatlantic
Council on Migration.