Stelzer on Immigration:
Some Lessons for Australia
by Jason Soon
Click here for PDF version

Economics-both through theoretical research and empirical studies-can pay an important role in formulating a rational immigration policy.

The purpose of this article is to flesh out some of the propositions made by Irwin Stelzer in the preceding article and their implications for Australia. As many of the relevant issues have already been identified by Stelzer, I will discuss these issues by setting them out under a series of headings and elaborating upon them.

The welfare economics of free migration

Stelzer argues that the free movement of labour resources, like that of goods and of capital resources, enhances efficiency. There is indeed some theoretical and empirical evidence for this proposition in the economic literature. To take just one example, Bob Hamilton and John Whalley estimate that gains to the world economy from removing immigration barriers could well be enormous and greatly exceed the gains from removing trade barriers; they calculate that these gains would more than double worldwide real income.1

There are a number of catches however.

First, most of the attributed gains from free migration of labour arise from lifting immigration controls which currently maintain the cost of labour in developed countries at a higher wage than would be the case if all workers in developing countries willing to work for less than the market wage in developed countries were able to migrate to developed countries and compete for jobs with 'locals'. Increased labour market competition from 'free migration' policies would consequently allow these gains from trade to be captured. The resulting benefits arise from the market wage being bid down in developed countries. This is not only to the benefit of native-owned firm and consumers who get to consume cheaper goods and services, but also immigrants from developing countries who enjoy greater economic opportunities compared to those that exist in their countries of origin. These benefits, however, accrue at the expense of native workers in developed countries who compete with immigrants.2 Though the size of the economic 'pie' may increase internationally, there is also a redistribution of income from native workers who compete with immigrant labour to those who use immigrants' services.3

In theory, as long as the gains from cheaper labour exceed the costs suffered by the native workers in competition with new migrants, immigration is economically beneficial because the 'losers' can be compensated through redistributive policies while still leaving a surplus to be enjoyed. In practice it is not so simple. It is highly unlikely that the majority of Australians would be prepared to accept this kind of tradeoff.4 Even if they were, existing policy settings will decrease the gains from such a policy. Current labour market regulations mean that some of the increased competition in the labour market will manifest itself as higher unemployment rather than lower labour costs. Other policy settings are also likely to dilute the main labour market benefits of free migration of labour and add to taxpayer costs, such as our welfare state.5

The second catch is that the gains discovered by these studies are gains to the international community as a whole. In other words, they include a measure of welfare gains for immigrants themselves, and for residents of developing as well as developed countries. Should we give more weight to native workers in developed economies who may be worse off because of increased labour market competition from immigrants when to an immigrant or refugee from an impoverished country, working for a few dollars an hour in Australia may well be a significant improvement to conditions faced in his or her homeland? Economics by itself has nothing to say about this. To the extent, however, that economics is being used to guide the policymaker, then a measure of welfare that equally weights the interests of foreigners with the interests of natives is unlikely to have much influence.

These considerations support the case for adopting Stelzer's approach of using a normative approach based solely on national self-interest; that is, of evaluating immigration from the perspective of wanting to maximise the wealth of the existing population. This sets a higher hurdle for increased immigration than any other approach. From this perspective, Stelzer is correct in arguing that the expected gains from free migration of labour would create a presumption towards a generous immigration policy but stop short of removing all immigration barriers.6

Labour market characteristics of immigrants

If a policy based on 'national interest'-that is, one that increases the welfare of existing residents-is to be our normative standard for evaluating immigration policy, then this suggests that an 'optimal' immigration policy should prefer immigrants with positive labour market characteristics such as employability and complementary skills which are in high demand but short supply in the Australian labour market.7

Available data suggests that the labour market characteristics of current migrants vary strongly according to the migration programme through which they entered Australia.

Generally, immigrants are more likely to secure employment if they: come from an English-speaking background (or have strong second-language English skills); have a recognised qualification or skill; and/or, have longer period of residence. Such persons often have lower measured unemployment rates than the non-immigrant population.

The 1995 Longitudinal Survey of Immigrants to Australia (LSIA) showed that, five to six months after arrival, of immigrants in the business migration stream, under 3% were unemployed, compared with 85% for humanitarian entrants, 39% for preferential family and 36% for concessional family migrants. Overall, one third of recently-arrived adult migrants depend on welfare payments soon after arrival and around 25% are still dependent in their second year here. However these figures must be put into the context that unemployment rates for other new labour market entrants are high as well. For instance, a 1995 survey of Australian first time labour market entrants (those who just finished the education system plus others) estimated their unemployment rate to be 45%.8

The 1995 Longitudinal Survey also revealed that labour market outcomes for immigrants tend to improve as their period of residence in Australia lengthens. While just 37% of immigrants were employed six months after arrival in Australia, this increased to 59% after three and a half years. Unemployment rates drop most quickly for economic migrants; by the end of their third year those who entered under the skilled (including concessional family) category have unemployment rates below the Australia-wide average.9

If anything, the labour market performance of immigrants is likely to improve further with recent policy changes. In 1997-98, a new Skill category-Skilled-Australian Linked-replaced the Concessional Family category. This saw Skill migration arrivals exceed Family stream arrivals in 1997, a trend which has since continued.10

There was another change to the selection process in July 1999 when the points test requirements relating to skill level, age and English fluency were strengthened, and increased emphasis was placed on targeting migrants with specific skills which are in demand in the Australian labour market.

One example of a well-targeted programme, in this case, one that aims at picking migrants with entrepreneurial skills of value to the economy, is the Business Migration programme. Business migrants had a high propensity to export, with 60% of them being involved in exporting within two years of arrival.11 Those business migrants who started a new commercial enterprise employed, on average, seven persons, while those who either purchased or bought into an existing firm employed, on average, 37 persons. Average employment for all businesses owned by business migrants (arriving in the two years under review) was just over 14 persons.12

Demand-side effects of immigration

Another factor to consider is the demand-side effects of immigration. Immigrants are consumers of products such as housing, food, household inclusions, and transport, education and health services. This induces increased investment, which expands the nation's capital stock, leading to additional employment opportunities. The capital inflows can be substantial-business migrants, on average, bring $219,000 per capita (in 1992/93 dollars) into the country, followed by skilled labour at $28,000) and the Independent category ($21,000), though family reunion and humanitarian/refugee immigrants bring relatively little capital with them. 13

Indeed, most evidence suggests that the positive demand effects of immigration in creating employment outweigh the supply side effects of greater competition for jobs.14

The postwar immigration intake up to 1980 has already added around 40% to the average annual growth rate of the production of goods and services. 15 One particularly intriguing study found not only that immigrant spending from past savings would increase the demand for labour and create job vacancies so that it has almost certainly increased the short-term employment probability of unemployed Australian residents but, paradoxically, this effect was more likely when the economy was weak.16

Immigration and the population structure

In his article, Stelzer mentions the dilemma of European policymakers who face a population reluctant to admit more immigrants, but in the absence of higher immigration, a growing number of retirees paid for by taxes of fewer young workers. This dilemma is one faced by all Western countries including Australia.17 Indeed, the ageing of the population was cited by the Department of Immigration and Multicultural Affairs (DIMA) as a reason for the increase in the immigration intake for 2001-2002 by 6,000 places to 85,000.

The ageing of the population has serious implications for the financing of government services. As birth and death rates have fallen in Australia, the speed and the future level of population ageing have increased sharply. There will be 100,000 fewer people each year at the young end of the age structure and almost 78,000 more people each year at the old end of the age structure. In 1998, just over 12% of Australia's population were aged 65 years and over. Population projections indicate that this will at least double in the next 40 years.18 An older population implies increased costs for aged pensions and health and aged care services.

The relevance of immigration to this issue is that, as a 1996 research paper concluded, large inflows of people with higher fertility and a younger age structure than the native population can retard the process of population ageing and therefore the growth in the ratio of social expenditure to GDP.19

The effect of immigration on ageing, however, is subject to diminishing returns. Between zero and 50,000 annual net migration, the 50,000 migrants would reduce the proportion of the population aged 65 years and over in the year 2098 by 3.0 percentage points. However, between 200,000 and 250,000 annual net migration, the additional 50,000 migrants would reduce the proportion aged 65 years and over by only 0.5 percentage points.20

Recent DIMA research has reinforced this finding. It concluded that given current trends in fertility and mortality, annual net migration to Australia of at least 80,000 persons is necessary to avoid spiralling population decline and substantial falls in the size of the labour force. Levels of annual net migration above 80,000, however, become increasingly ineffective in the retardation of ageing. 21 This conclusion is hardly surprising, considering that migrants age like everyone else. In any case, Australia is better equipped than other countries (through, for instance, its superannuation policies which essentially involve the current generation of workers being compelled to fund their own retirement) to deal with the contingency of an ageing population.

Immigration, government spending and taxes

The idea that immigration is a desirable policy for tackling the problem of an ageing population and the higher social expenditures that come with it presumes that immigrants make a net contribution to taxes rather than being a fiscal burden. Similarly the issue of 'how much immigrants cost' in terms of the demands they place on government services will be relevant to the design of an immigration policy that is optimised to enhance the welfare of current residents. How then do the sums add up?

While it is true that immigrants add to the demands placed on government services, there may well be economies of scale both in private and public infrastructure from the increased population created by immigration. To date there has been no precise quantification of these possible economies of scale.

What can be discerned is that, as is true with the case of labour market characteristics, the effect of migrants on government outlays depends on the migration programme through which they entered Australia. Across the migration categories, outlays tend to peak in the third year and then decline to a plateau. Receipts, on the other hand, tend to rise steadily throughout the five years.22

The most costly migrant category is the humanitarian stream (mainly for refugees). Outlays on the humanitarian category are significantly greater than any other category, but this rapidly falls over the second and third years and then plateaus. These outlays are minimally offset by receipts.

The preferential family category (better known as 'Family reunion' migrants) shows characteristics different from each of the other categories, as the net budget impact does not steadily improve over the five years as occurs in all other categories. The net impact improves over the fourth and fifth years.

The highest receipts categories of migrants are the independent and business categories (these are both Skilled migrant categories). Independents begin solidly and have nearly doubled their receipts by the fifth year. The Business category begins with high receipts and in the fifth year has increased those receipts although to a lesser extent than the independent category.

Nonetheless, on a lifetime contributions basis, overseas-born of all durations in Australia are more favourable to the public purse than Australian-born. This is not surprising considering that Australia does not have to pay for their childhood education and health costs. The net present-value cost to the Federal Budget in the early 1990s for a 23 year old born in Australia was $223,000, well ahead of the $13,000 for an immigrant of the same age.23

Immigration and the natural environment

One additional factor to consider is the issue of whether there are population carrying capacity limits to the amount of immigration that Australia can absorb from an ecologically sustainable perspective.

Unfortunately, though a lot of figures have been bandied around, there is little strong support or consensus for any of them.24 No attempts at quantifying an environmentally sustainable population can properly take into account future technological improvements and advances in environmental management that may increase carrying capacity.

More problematically, despite the attempts by low population growth activists to invest these figures with significance, the welfare implications of exceeding an alleged 'ecologically sustainable population' are unclear. For example, a study of Australia's carrying capacity in 1994 concluded that our supplies of water, fibre and electricity are adequate for a population of 100 million or more while our land-based food production capacity is enough for a population of 45 million people.25 The same study found that we already exceed our carrying capacity in timber, fisheries and oil. Obviously, despite this we are not worse off than we were in 1994. None of these facts points clearly towards the optimum population size, since international trade can redress any insufficiencies.

Pricing versus quotas in immigration policy

Stelzer argues that, at least with respect to the skilled migration programme, greater reliance could be placed on market forces by replacing the 'points' system for selecting migrants with a system of allocating available visas to the highest bidders.

One possible merit of Stelzer's proposal is that in principle it might be more informationally efficient. This is because the current 'points' system for selecting skilled migrants according to their certified skills and abilities is essentially a tool for ensuring that immigrants selected will be as valuable as possible to the economy rather than being a net burden. This is consistent with the objective of an immigration policy that maximises the welfare of current residents. A price-based system could achieve the same end without the administrative cost and error cost of micromanaging the selection process involved in a points-system.

First, the need to pay for a visa will increase the high-degree of self-selection. To put it bluntly, a price-based system would cut down even further on the number of 'slackers' tempted to migrate to Australia and self-select for those most likely to be economically productive. Presumably those likely to take up such an offer would also think twice about migrating if their skills are not in demand so the self-selection process would render a skills-test redundant.26

Second, the government would be assured of some benefit whatever happens: it receives revenues from the entrance fee charged in addition to the revenues which it has a high probability of getting if the most economically productive migrants are self-selected to apply. Indeed, on a politically pragmatic level, this may be an effective means of 'selling' immigration to the public. To placate those worried about the need for an ecologically sustainable population, part of the proceeds of such a scheme could be put into a trust fund for environmental management.

It might be argued that any entrance fee, if imposed in the past, would have excluded current immigrants who have nonetheless turned out to be economically productive, the implication being that such an approach would be over-exclusive. This objection could be addressed by allowing the immigrant to decide how he or she wants to pay the fee. The fee could be paid upfront. Alternatively, it could take the form of a discriminatory income tax, withheld from income earned after immigration with obligations to repay rising as the immigrant's income rises, much like the Higher Education Contribution Scheme (HECS) for university students.

More realistically, even if some sort of fee-based system were introduced merely as a supplement to the current immigration quota, it would have the effect of diverting some of the profits from the people-smuggling trade to the Australian government. For instance, if an additional 10,000 people (over the current quota) were admitted on this basis, this represents a potential 10,000 people who might have paid money to people-smuggling rings instead.

Conclusion

Immigration policy is an area that intersects with many others spheres of policy including environmental management, population policy, skills and training, infrastructure and urban planning and, unfortunately, since the terror attacks on the US of September 11, issues of national security. As Stelzer's paper demonstrates, and as these comments attempt to illustrate, economics-both through theoretical research and empirical studies-can have an important role to play in formulating a rational immigration policy which by necessity has to be one which properly accommodates all these competing needs and priorities.

Endnotes

1 Bob Hamilton and John Whalley, 'Efficiency and Distributional Implications of Global Restrictions on Labour Mobility', Journal of Developmental Economics 14 (1984), 61. Howard Chang, 'Migration as International Trade: The Economic Gains From the Liberalised Movement of Labour', UCLA Journal of International Law and Foreign Affairs, vol. 3 (Fall/Winter 1998-99), 371-414, summarises the economic arguments for free migration of labour. See also Paul Levine, 'The Welfare Economics of Immigration Control', Journal of Population Economics 12:1 (1999), 23.
2 The negative side to this tradeoff is discussed from a US perspective in George Borjas 'The New Economics of Immigration', The Atlantic Monthly (November 1996).
3 Though this labour market effect is indeed an important theoretical reason for the calculated gains from free migration of labour, in practice other factors like demand-side factors may outweigh this effect so that there is no substantial measured impact on the wages of natives competing with immigrants. For instance Maria Enchautegui, 'Effects of Immigrants on the 1980-1990 US Wage Experience', Contemporary Economic Policy 13:3 (1995), 20, finds that controlling for personal characteristics, the hourly wages of the average native and immigrant worker in areas of high and medium immigration relative to areas of low immigration increased between 1980 and 1990. For another sceptical note on this which may reflect other factors outweighing the immediate effects of tighter competition in the labour market from increased immigration, see Rachel Friedberg and Jennifer Hunt, 'The Impact of Immigrants on Host Country Wages, Employment and Growth', Journal of Economic Perspectives 9:2 (1995), 23. They find that a 10% increase in the fraction of immigrants in the population reduces native wages by only 0-1%.
4 Indeed their strong unwillingness to accept such a tradeoff was arguably one of the economic motivations behind the so-called 'White Australia' policy.
5 As we shall see, arguments about the possible welfare dependency and unemployment of migrants adding substantially to the costs of immigration do not in fact apply, at least with respect to current and possibly higher intakes, but this does not mean that there is not a limit at which they would start to bite into the benefits of a freer flow of labour across countries.
6 ÊIn any case these full calculated gains would not apply with a unilateral dismantling of barriers to free migration.
7 Friedberg and Hunt, 'The Impact of Immigrants on Host Country Wages', 23, find no large adverse impact of immigration on natives' wages concludes that impact on natives' per capita income growth depends crucially on the immigrants' human capital levels.
8 This is cited at p. 18 of Rainer Winkelman, 'Immigration Policies and Their Impact: The Case of New Zealand and Australia', Discussion Paper No 169 (Institute for the Study of Labour, 2000).
9 See also Australian Bureau of Statistics, Migration 1999-2000, ABS Cat No. 3412.0 (Canberra: ABS, 2000).
10 Australia's trend towards rising skill levels for successive cohorts of immigrants as a result of changes in government policy over time is different from that of the US where the research consensus has been that the quality and skills of successive immigrants have declined over time. See Richard Fry 'Has the Quality of Immigrants Declined? Evidence From the Labour Market Attachment of Immigrants and Natives', Contemporary Economic Policy 14:3 (1996), 53.
11 Australian Chamber of Commerce and Industry, 'Australia's Immigration: Policies and Programs' (October 1998). For a Canadian perspective on the link between immigration and trade opportunities, see Keith Head and John Ries, 'Immigration and Trade Creation: Econometric Evidence from Canada', Canadian Journal of Economics 31: 1 (1998), 47, which found that a 10% increase in immigrants is associated with a 1% increase in Canadian exports to the immigrant's home country and a 3% increase in imports.
12 Australian Chamber of Commerce and Industry, 'Australia's Immigration: Policies and Programs'.
13 Australian Chamber of Commerce and Industry, 'Australia's Immigration: Policies and Programs'.
14 See for example D.S. Harrison, 'The Impact of Immigration on a Depressed Labour Market: The South Australian Experience, The Economic Record 60: 168 (1998), 7-67, and more recently Bruce Chapman and Deborah Cobb-Clark, 'A Comparative Static Model of the Relationship Between Immigration and the Short-Run Job Prospects of Unemployed Residents', The Economic Record, (2000), available at <http://papers.ssrn.com/sol3/delivery.cfm/99052610.pdf?abstractid=166292> for evidence from Australia. See T. J. Samuel and T. Conyers 'The Employment Effects of Immigration: A Balance Sheet Approach', International Migration 25 (1987), 283-90, for Canadian evidence.
15 Australian Chamber of Commerce and Industry 'Australia's Immigration'.
16 Chapman and Cobb-Clark, 'A Comparative Static Model of the Relationship between Immigration and the Short-Run Job Prospects of Unemployed Residents'.
17 For a recent US perspective on this, see Kjetil Storesletten, 'Sustaining Fiscal Policy Through Immigration', Journal of Political Economy 108:2 (2000), 300.
18 Peter McDonald and Rebecca Kippen, 'The Impact of Immigration on the Ageing of Australia's Population', (Canberra: Department of Immigration and Multicultural Affairs, 1999), available at <http://www.immi.gov.au/population/ageing.htm>
19 J. Alvarado, and J. Creedy, Migration, Population Ageing and Social Expenditure in Australia (Canberra: Department of Immigration and Multicultural Affairs, 1996).
20 C. Young, Australia's Ageing Population-Policy Options (Melbourne: Bureau of Immigration Research, 1990).
21 McDonald and Kippen, 'The Impact of Immigration on the Ageing of Australia's Population'.
22 ACIL Consulting, 'Impact of Migrants on the Commonwealth Budget', A Report to the Department of Immigration and Multicultural Affairs (June 1999), available at <http://www.immi.gov.au/population/acil/acil1.htm>
23 Australian Chamber of Commerce and Industry, 'Australia's Immigration'.
24 See Gavin W. Jones 'An Australian Population Policy', Australian Parliamentary Research Paper 17 (Canberra, 1996-97) for a summary of these studies.
25 Ian Lowe, Understanding Australia's Population Debate (Canberra: Department of Immigration and Multicultural Affairs, 1996).
26 To be perfectly effective, a points test would have to be adjusted every so often according to changing conditions. The rational incentive to properly 'scope out' one's investment in a visa to Australia may well be a sufficient substitute for any points test

Author
Jason Soon is an economist and consultant at Network economics Consulting Group, and Adjunct Scholar with CIS. The views expressed here are those of the author.

Policy is the quarterly review of The Centre for Independent Studies. For more information on subscribing to Policy, click HERE

If you are interested in the Centre's activities and publications, why not subscribe to e-PreCIS, our regular email update on the latest news and events.

(e-PreCIS requires html capable email facilities, such as Microsoft Outlook Express or Netscape Messenger)