The Economic Goals of Canada’s Immigration Policy: Past...

The Economic Goals of Canada’s Immigration Policy: Past and Present425

CANADIAN PUBLIC POLICY – ANALYSE DE POLITIQUES, VOL. XXV , NO. 4 1999

The Economic Goals of Canada’sImmigration Policy: Past and PresentALAN G. GREEN

Department of EconomicsQueen’s UniversityKingston, Ontario

DAVID A. GREEN

Department of EconomicsUniversity of British ColumbiaVancouver, British Columbia

Nous examinons les objectifs économiques de la politique actuelle d’immigration, ainsi que le rôle de l’immi-gration dans l’ensemble des politiques économiques. Nous commençons par décrire les objectifs écono-miques de la politique d’immigration au cours du dernier siècle. Nous décrivons ensuite les éléments cibléséconomiquement par la politique d’immigration en les reliants aux tendances historiques. Finalement, nousexaminons un ensemble d’objectifs économiques potentiels pour la politique d’immigration qui sont suggé-rés par l’histoire politique canadienne. En conclusion, il apert que les objectifs économiques ne devraientpas constituer l’orientation de base de la politique d’immigration dans un avenir immédiat, car d’autrespolitiques mieux adaptées peuvent atteindre ces objectifs. L’immigration devrait continuer d’être un élémentde base de notre structure sociale.

We examine the economic goals of current immigration policy and what role immigration should play inoverall economic policy. We proceed by describing the economic goals of immigration policy throughoutthis century. We then describe current economically targeted elements of immigration policy and relatethem to historical trends. Finally, we examine a set of potential economic goals for immigration suggestedby Canada’s policy history. We conclude that economic goals should not form the defining orientation ofimmigration policy in the near future since other policies are better situated to meet those goals. Immigra-tion should continue as a defining element in our social fabric.

INTRODUCTION

Canadian immigration policy is currently the fo-cus of considerable debate and is in the pro-

cess of being reviewed by the federal government.The key question in the debate is what should bethe goals of our immigration policy. Once that ques-tion is answered, other questions such as how to setthe size and composition of the inflow should, hope-

fully, be easier to answer. A variety of potential goalspresent themselves: humanitarian goals such as help-ing refugees and reuniting families in Canada, so-cial goals such as altering the nature of Canadiansociety, and goals related to foreign policy. ForCanada, though, one of the main uses of immigra-tion policy in the past has been as a tool of eco-nomic policy. In this paper, we examine the eco-nomic goals of current immigration policy and the

426 Alan G. Green and David A. Green


role immigration should play in overall economicpolicy. To do this, we argue that one must considercurrent and potential goals in the context of the his-toric goals of immigration policy. This is true, inpart, because the debate on current immigrationpolicy often refers to historic precedent as a sourceof justification for particular options. This usage ofhistorical argument reflects the fact that for a policysuch as immigration where the potential effects areboth large and subtle, the main basis for understand-ing its impact must be in our own history. Thus, webelieve that understanding how immigration policyhas been used in the past is a necessary precursor todiscussing how we want to use it in the present andfuture. Further, knowledge of past goals of immi-gration policy allows us to understand whether cur-rent and potential policies are radical shifts fromwhat has been deemed economically and politicallyfeasible in the past. To the extent that new policiesrepresent radical changes, arguments in their favourbear a greater burden of proof.

We proceed through our examination in threesteps. First, we describe the economic goals pur-sued by immigration policy from the earliest yearsof this century to the present. Second, we describecurrent economically targeted elements of immigra-tion policy and relate them to the historical trends.Third, we examine a set of potential economic goalsfor immigration policy suggested by Canada’s policyhistory. Our aim is not to determine whether eachpotential goal is reasonable in itself but to decidewhether immigration is the best means of attainingthat goal.

We argue, first, that Canada’s immigration policyhistory up to the late 1980s can be summarized asalternating periods of large inflows targeted at spe-cific economic goals and periods of virtual shutdown of immigration in the face of poor domesticlabour market conditions. Most importantly, eco-nomic goals formed a central orientation of immi-gration policy throughout much of the twentiethcentury. Second, recent policy departs sharply fromour history in not substantially reducing inflows

despite high and persistent unemployment rates. Thespecific economic goals of the policy are also notwell specified, in contrast to the past. This raisesthe question of whether immigration policy shouldreturn to its focus on economic goals and, if so, onwhich goals. Using a combination of current immi-gration research and comparisons to historical pat-terns, we do not find a strong argument in favour ofusing immigration for virtually any of a consider-able list of potential economic goals. Our argumentis less that immigration is not useful for such goalsas building up the stock of human capital in theeconomy than that other policies are superior formeeting those goals. Our conclusion is that eco-nomic goals should not form the defining orienta-tion of immigration policy as we move into the nextcentury. That does not mean, however, that immi-gration should be stopped. There is also no clearevidence of large economic costs to immigration, atleast as long as it is well managed. Thus, the bestimmigration policy for Canada appears to us to beone that focuses on humanitarian goals while pay-ing attention to short-term costs of immigration inthe Canadian labour market. Immigration in the last20 years has played a significant role in definingCanada as a country with a richly diversified cul-ture and, through refugee policy, as a generous coun-try. Those should continue to be its main goals.

Finally, we do not pretend to do an adequate jobof placing immigration in the context of political,social, and cultural trends. Our aim is to discussimmigration as an economic policy. We stronglyencourage readers to seek out work by authors suchas Abella and Troper (1983); Avery (1979); Iacovetta(1992); Ramirez and Del Balso (1980); and Rich-mond (1967) in order to understand the importanceof immigration to Canada in a wider context.

THE HISTORIC GOALS OF CANADIAN

IMMIGRATION POLICY

In this section, we describe the main economic goalsaddressed by immigration policy over Canada’s



history. We describe the goals within each of a setof periods, where the period endpoints are definedby major changes in the stated goals of the policy.We do not attempt to assess whether the policy wassuccessful in meeting each of these goals. We aremore concerned with how immigration has been per-ceived as an economic policy tool and how that con-ditions our current options.

Before beginning on our investigation, we pro-vide some definitions. It will prove useful to followFreda Hawkins (1972) in dividing the economicgoals of immigration policy into short- and long-term goals. We will categorize achieving demo-graphic benefits (those relating to size and age com-position of the inflow), benefits from general im-migrant characteristics such as flexibility, and in-vestment and trade flow benefits as long-term goalsof immigration policy. Achieving benefits frommeeting immediate shortages in the labour market,we will categorize as short-term goals. As we shallsee, the setting of Canadian immigration policy hasoften reflected a basic tension between using immi-gration as a tool for promoting long-term growthand using it as part of short-term labour marketpolicy. However, the two sets of goals are not al-ways in conflict and in fact overlap at times.

The pursuit of these various goals is conditionedby attempts to minimize the potential costs of im-migration. Thus, a defining feature of Canadianimmigration policy has been a concern that the in-flows not exceed the “absorptive capacity” of theeconomy. According to this concept, the ideal rateof absorption depends on the ability of the economyto provide employment for new immigrants at theprevailing nominal wage (see Timlin 1960). Hence,in periods of rising unemployment the absorptivecapacity for new immigrants declines and the gov-ernment takes steps to limit the number of arrivals.This policy is reversed as the domestic labour mar-ket tightens. This is a recognition that large immi-grant inflows may displace native-born workers inthe host economy in the short run even if there areoffsetting long-run benefits from the inflows. In-

deed, the concept of absorptive capacity is probablymore strongly related to short-term than long-termgoals: if the government is using immigration tomeet excess demand for specific occupations thenby definition the absorptive capacity of the economyis being considered. Similarly, if the government ispursuing primarily longer term goals then it is lesslikely to respond to short-term cost considerations.

1870-1913This was the period which, after a slow start duringthe last decades of the nineteenth century, saw thesettlement of the west, high levels of investment,rapid economic growth and the establishment of anational economy. During these years, and indeedup to 1930, immigration policy was part of a gen-eral set of national policies. These included the com-pletion of three transcontinental railways, the im-position of high levels of protection on the importof secondary manufactured goods, and the adoptionof a land policy aimed at inducing immigrants tosettle in the west. As a set, these policies were meantto tie Canada together into an integrated whole witha strong eastern manufacturing sector selling itswares to an expanding western resource sector. Thepromotion of immigration was clearly an importantelement in this development strategy. Faced withlarge immigration inflows from the United Statesand overseas starting in 1896, the government wasforced to replace the original Act of 1869 with anew Act in 1910. The new Act included a basic ap-proach of focusing on a prospective immigrant’scountry of origin; an approach that was unchangeduntil a non-discriminatory set of regulations wascreated in 1962.

The stated goal of immigration policy up to WorldWar I was to secure farmers, farm workers, and fe-male domestics. The search for such workers wasto be concentrated in Britain, the United States, andnorthwestern Europe. In actual fact, despite the pub-lished goals of seeking only farmers for the west,the actual distribution of immigrants was aboutequally divided among those intending to work inagriculture, manufacturing, and the service sector.



Moreover, immigrants were destined to all regionsof Canada, not just the west (Green and Green1993).1 Further, when excess demand for labour in-creased in this first period, the government wasforced to abandon its racist goals to some extent andexpand recruitment beyond the traditional sourcecountries. Specifically, this meant bringing in im-migrants from Central and Eastern Europe (Avery1979).

1919-1929During this decade the official immigration regula-tory structure was put in place. The first majorchange came with the 1919 revisions to the 1910Immigration Act. These revisions established a lit-eracy test for all prospective immigrants; expandedthe terms of section 38 of the 1910 Act to allow thegovernment, through Order-in-Council, to limit thenumber of immigrants admitted in a given periodand to refuse admission due to conditions “tempo-rarily existing in Canada”; and added the word “na-tionality” to that of “race” to define the origin ofimmigrants. The net effect of these changes was toexpand the power of the government over the level,timing, and ethnic composition of immigration.

This period witnessed the first official divisionof source countries into preferred and non-preferredgroups; a division that was to form the core of Ca-nadian immigration policy until the early 1960s.Admission from the preferred countries was basedsolely on country of origin.2 Prospective immigrantsfrom non-preferred countries were admitted undera variety of conditions. Applicants from northernand western Europe were treated as almost equal tothose from preferred countries, while those fromcentral, eastern, and southern Europe faced stricterregulations. Immigrants from other regions wereadmitted only if sponsored by a relative already le-gally admitted to Canada.

In 1925, just two years after the formal divisionof countries into these two categories, the govern-ment passed the Railways Agreement, chartering thetwo major railway companies (the CPR and the

CNR) to, “recruit, transport and place in Canada,agricultural families, farm labour and domestic serv-ants.” The agreement limited their search for immi-grants to central and eastern Europe. The pressureto extend the search into non-preferred countries inthis way arose from the scarcity of farm labour intraditional source countries (e.g., Britain and the US)coupled with growing demands for such workersfrom the west. Thus, supply considerations forcedthe government to alter its official goals. Recentwork (Green 1994) indicates that this policy of steer-ing immigrants to the west was highly successful.

The period just following World War I is also thefirst time in which the government formally ac-knowledged the concept of short-run absorptive ca-pacity. In 1918, the government established the “Em-ployment Service Council” — a federal body witha mandate to control the level of immigration. Theconcept of absorptive capacity was first imple-mented in 1921 when the government refused, inthe face of rising unemployment, to issue applica-tions to employers wishing to bring in foreign work-ers. This, combined with the Railway Agreement,signalled a move toward active intervention in theselection of immigrants by the government, even ifit used other agents for actual implementation. Steer-ing immigrants to specific sectors and regions waspart of the stated government policy from the latenineteenth century but apparently only became a factafter 1919.

The 1930s and 1940sAfter almost six decades of actively recruiting im-migrants, the door closed to most newcomers withthe passage of Order-in-Council PC 695 on21 March 1931 and remained that way until afterthe end of the Second World War. Figure 1, in whichwe plot immigration as a percentage of the currentpopulation over time, shows that immigration fellto near zero as a percentage of the population of thetime.3 Immigration was closed off from the wholeworld, with the exception of Britain and the US, andwith the occupation exception of farmers with capi-tal. Even in these difficult times family reunification



was a priority, although this right was narrowly lim-ited to include only immediate family members whowould not directly enter the labour force.

The new regulations were a reaction to extremeeconomic conditions. By 1931 the unemploymentrate stood at over 11 percent and, although the offi-cials of the time did not know how long these con-ditions would prevail, the situation was such that apolicy shift of this order of magnitude seemed jus-tified. Nation-building, which had been of key im-portance to the government from 1870 to 1930, hadbeen replaced by preserving the nation in the faceof this major depression. This was then followed bya restrictive policy dictated by the exigencies of war.

1946-1962Immigration policy in the immediate postwar pe-riod was dominated by two main events. The firstwas the large influx of displaced persons from

Europe. The second was the establishment of clearethnic and economic goals for immigration policy.The displaced person flow fit with the broad ethnictargets, since most of those accepted were from Eu-rope, but was sometimes in conflict with the eco-nomic goals.

The tone of immigration regulations for the dec-ade and a half following the end of hostilities wasset by Mackenzie King’s statement before the Housein May 1947. It is worth repeating in some detailsince it was a blueprint for government policy onthis issue.

The policy of the government is to foster thegrowth of the population of Canada by the en-couragement of immigration. The governmentwill seek legislation, regulation and vigorousadministration, to ensure the careful selection andpermanent sett lement of such numbers of

FIGURE 1Immigration as a Percentage of the Current Population, Canada, 1896-1997

5.5

1905 1915

5.0

4.5

4.0

3.5

3.0

Per

cent

Year

1925 1935 1945 1955 1965 1975 1985 1995 20050.0

0.5

1.0

1.5

2.0

2.5



immigrants as can advantageously be absorbedin our national economy.... There will, I am sure,be general agreement with the view that thepeople of Canada do not wish, as a result of massimmigration, to make a fundamental alterationin the character of our population. Large-scaleimmigration from the Orient would change thefundamental composition of the Canadian popu-lation (Quoted in Green 1976, p. 21).

The statement has six main components. First, im-migration was to be used to promote populationgrowth. Second, immigration would improve thestandard of living of the extant population. Thiswould operate by enlarging the domestic market, thatis, by promoting economies of scale. Third, immi-gration was to be selective. Fourth, immigration wasto be related to the absorptive capacity of theeconomy. Fifth, immigration was to be a matter ofdomestic policy, that is, national concerns were para-mount. Sixth, immigration should not change thebasic character of the Canadian population, mean-ing that restrictions on Asian immigration must re-main in place (see Hawkins 1972).

The statement contained a potential tension be-tween the main goals of immigration policy. It en-compassed long-run benefits (i.e., population growthand economic development), while reaffirming thegovernment’s commitment to matching short-runlabour market conditions. As in the 1920s, no realtension existed in the immediate postwar periodbecause the desired immigrant was essentially un-skilled. The policy was dominated by high demandfor labourers in the booming resource sector, par-ticularly in forestry and mining. Thus, the govern-ment could increase the population size and meetperceived labour shortages at the same time(Hawkins 1972). However, supply constraints fromtraditional source countries meant that the govern-ment had to partially abandon the racist element ofthe policy and extend its search to non-preferredcountries (Richmond 1967). It did this in part byrelaxing conditions of acceptance for immigrantsfrom non-preferred countries and partly by expand-

ing the sponsorship rights of immigrants from central,eastern, and southern Europe to parallel those avail-able to immigrants from traditional source countries.

The extension of sponsorship rights to landedimmigrants from non-traditional sources in Europecreated difficulties. Landed immigrants from south-ern Europe, and Italy in particular, were much morelikely to make use of sponsorship rights than immi-grants from traditional source countries. The resultwas a much more unskilled inflow beginning in thelate 1950s (Green and Green 1995). This occurredjust at a time of economic policy change as thegovernment sought to steer the economy away froma resource base and toward a modern manufactur-ing structure. As part of this change in direction, itwas argued that Canada needed skilled workers andneeded them immediately. To accommodate longerterm skill requirements, the postsecondary educa-tion system was to be expanded, but the more im-mediate needs would have to be met throughimmigration.4

The contrast between the large, unskilled inflowand the goal of raising the skill level of theworkforce created the first substantial tension be-tween the short- and long-term goals of immigra-tion policy. This tension had a direct institutionalembodiment in the two departments dealing withimmigrants in this period. The Department of Citi-zenship and Immigration took a longer run view,while the Department of Labour attempted to tie thelevel of immigration to the business cycle and,moreover, to specific occupational job vacancies.Hawkins (1972) argues that, institutionally, the late1950s can be seen in terms of a battle between thesetwo departments. From the longer term view of Citi-zenship and Immigration, a larger, less skilled in-flow was not necessarily a troubling outcome. How-ever, for the Department of Labour, it was a disas-ter, especially coming as it did at a time of risingunemployment.

The justification for immigration as a means offilling skill gaps is significant since it recurs many



times in the next three decades.5 This approachpoints to an implicit input/output model of theeconomy in which production is hampered if oneimportant factor is missing. The idea of using im-migration to prevent that problem is at the heart ofthe selection system created in the next period weconsider. It did, however, increase potential supplyproblems since the world supply of skilled workerswas less than that of unskilled workers.

1962-1973In 1962 the government abandoned its long-standingracist immigration policy. Henceforth, admissionwas to be based on an individual’s personal charac-teristics, especially their education and other skillsqualifications, rather than his or her nationality. Thisis clearly set out in the 1962 policy statement bythe minister of immigration and citizenship, EllenFairclough:

The key to our immigration policy will be theconsistent application of proper selection stan-dards designed to bring the best possible settlersto Canada. I am sure all Canadians agree that oncethese standards are established they should be ap-plied consistently to all who seek admission tothis country, except where the admission of theimmigrant is based on compassionate grounds oron close relationships (Quoted in Green 1976,p. 37).

The lack of mention of population growth as a maingoal is in striking contrast to Mackenzie King’s 1947statement. The emphasis on selection standards in-dicates a victory for the short-run view of immigra-tion policy championed in the Department of La-bour. Indeed, even the elimination of the racist biasin the policy could be seen cynically as an attemptto expand the search area in response to inadequatesupplies of skilled immigrants from traditionalsources. This victory was consolidated with theamalgamation of the Department of Citizenship andImmigration and the Department of Labour into theDepartment of Manpower and Immigration in 1966.The new department was primarily concerned with

linking the level and composition of immigration tothe immediate needs of the domestic labour market(see Hawkins 1972, pp. 127-31).

The shift in policy was not without complications.Eliminating racism in selection meant that immi-grants from all parts of the world gained greatersponsorship rights. As immigrants from countrieswith lower average education levels exercised theserights, the result could have been an increasinglyunskilled inflow. Also, the new selection standardswere not well specified, implying that considerablediscretionary power was being placed in the handsof overseas immigration officers.

The solution came in 1967 with the introductionof the point system (Order-in-Council, 1 October1967, PC # 1616). The point system provided anobjective scale based on education, age, language,etc. against which applicants for admission couldbe assessed. It was the first major step to limit thediscretionary powers of immigration officers and toprovide them with a set of explicit guidelines.Applicants were divided into three main entryclasses: independents, whose admission dependedsolely on an assessment under the point system;nominated relatives, who were assessed under thepoint system, but were given bonus points based onfamily ties; and the family class, who were admit-ted based solely on kinship ties. The family classwas to be given top processing priority. With a totalnumber of immigrants roughly fixed by the size ofthe budget allocated to processing applications, thismeant that assessed immigrants were effectively aresidual under the new system. This was not an im-mediate issue, as the assessed part of the inflowmade up over 70 percent of immigrants in the re-mainder of this period (Wright and Maxim 1993).This feature was to become very important, how-ever, when the level of immigration was cut in therecessions ahead.

The form of the original point system indicatessuccess for those who viewed immigration as animmediate labour market policy. The points assigned



to specific occupations were to be kept under con-stant review and changed as new information onvacancies became available. In fact the changes inpoint levels assigned to specific occupations weredistributed to overseas agents on a quarterly basis(Green and Green 1995). This was the ultimate vic-tory of the belief that immigration policy could bemicro-managed. Table 1 provides the distribution ofpoints across assessed categories in versions of thepoint system from 1967 to today. In 1967, an appli-cant needed 50 out of a possible 100 points to beadmissible. Of the total possible points, 40 percentwere based on assessment of predicted short-termsuccess in the applicant’s intended occupation anddestination.6 This was balanced by longer term ad-justment factors, such as education, implying thatthe short-term point of view had not won an abso-lute victory.

1974-1985This was a period of big swings in the business cycle.Immigration inflows were adjusted accordingly.Immigration fell between 1974 and 1978, rose from1978 to 1980 and then fell from 1980 to 1986. Themajor swings in the level of business activity in thisperiod were the first real test of the efficacy of thenew regulations governing admission. In some waysthey failed since applicants could obtain enoughpoints based on years of schooling, age, and lan-guage facility to meet or exceed the minimum re-quired for admission even though their specific skillsmight be in excess supply at the time. This failureindicates the difficulty of trying to control both thelevel and composition of the skilled component ofthe inflow with points assessed for personal char-acteristics. The government responded by imposing,in 1974, a ten-unit penalty in points assessment ifthe applicant did not have previously arranged em-ployment. The system, however, was successful inreducing the number of unskilled workers admittedto Canada (Green 1995; and Green and Green 1995).

Although the regulations established in the 1960sremained largely intact during this next period, thegovernment did bring in a new Immigration Act on

10 April 1978. This Act defined three main goals ofimmigration policy: (i) to facilitate the reunion inCanada of Canadian residents with close familymembers from abroad; (ii) to fulfil Canada’s legalobligations with respect to refugees and uphold itshumanitarian traditions; (iii) to foster the develop-ment of a strong and viable economy in all regionsof Canada. Under these provisions family membersand refugees were given top processing priority andCanada was committed to bringing in a substantialnumber of refugees every year, rather than just inemergency situations. This forced a reduction in theshare of immigrants who were assessed and repre-sented a shift away from a policy focused on eco-nomic goals.

Concerns over meeting short-term labour marketgoals and over absorptive capacity did not disap-pear, however. The first concern was met by focus-ing on applicants with arranged employment or thosewho could be directed to specific occupations. Ab-sorptive capacity concerns were met by altering thesize of the inflow in response to cyclical fluctua-tions. The largest adjustments occurred in May 1982when the government declared that the only inde-pendent applicants who could enter were those witharranged employment. At the same time, the gov-ernment announced substantial cuts in the level ofthe inflow which were implemented in the follow-ing year and remained in place for the following fouryears. These adjustments indicate that short-termeconomic goals still held precedence in this period.Nonetheless, economic goals took a back seat tohumanitarian goals.

1986-1993When the immigration door reopened after the re-cession of the early 1980s, it did so on a very dif-ferent basis than when it had been shut. In 1985,the new Conservative government undertook a re-view of immigration policy, the conclusions fromwhich are evident in a special report to Parliamentin June of 1985 (Canada1985a) and in the AnnualReport to Parliament on Future Immigration Levelsseveral months later (Canada 1985b). A central



TABLE 1The Points System Over Time

Factor ’67 ’74 ’78 ’86 ’93 ’96

Education 20 20 12 12 14 21

Experience - - 8 8 8 9

Specific vocational 10 10 15 15 16 -preparation

Occupational demand 15 15 15 10 10 -

Labour market balance - - - - - 10

Age 10 10 10 10 10 13

Arranged employment 10 10 10 10 10 4or designated occ.

Language 10 10 10 15 14 21

Personal suitability 15 15 10 10 10 17

Levels - - - 10 8 -

Relative 0/3/5+ 0/3/5 5 - - 5

Destination 5 5 5 - - -

Total 100 100 100 100 100 100

Pass Mark 50 50 50 70 67 *

Bars on Entry:1967 – no one category result can be conclusive either way.Feb. 1974 – applicant must either receive at least one unit for occupational demand or get ten points for arrangedemployment or designated occupation.Oct. 1974 – ten points would be deducted unless the applicant showed evidence of arranged employment or was goingto a designated occupation. Removed in April 1979 but reimposed in September 1979.May 1982 – only applicants with arranged employment are eligible for admission. Removed in January 1986.1992 – zero units for experience is an automatic bar unless the person has arranged employment.

Note: The table shows maximum points possible in each category. Maximum points and Pass Mark have been rescaledin 1993 and 1996 to put the system in terms of points out of 100.+ Points awarded depend on relationship to sponsor.* The Pass Mark varies by skill group.The total available points actually equals 74. The pass marks are: professional,52; skilled administrator, 52; technical, 47; trades, 45.



concern in these reports is that fertility in Canadahad fallen below replacement levels, with the im-plication that the population would begin to declinejust after the turn of the century if immigration wasmaintained at its then current low levels. The re-ports also reflect a belief that the economic compo-nent of the inflow should be increased. But, cau-tions the Annual Report, “the increase in the eco-nomic stream must not ... be at the expense of thesocial and humanitarian streams” (ibid.). Immigra-tion levels would be increased substantially and re-strictions on the economic component would belifted to try to make sure that component grew withthe overall levels, but the assessed inflow was stillvery much a residual. The new policy represented avictory for those who saw immigration as a long-run policy tool. Immigration, as an economic policy,would be used primarily to bolster populationgrowth and to readjust the age structure of the popu-lation so that there would be enough workers to payfor the baby boomers’ pensions and health care.

The outcome of these policy deliberations was aremoval of the prerequisite of arranged employmentfor independent applicants in January 1986. Thelevel of the inflow jumped from 83,402 in 1985 to99,219 in 1986, and then to 152,098 in 1987. Thesewere the first in a series of increases that eventuallyled to a level of nearly 250,000 in 1993.

The new policy also saw the expansion of thebusiness component of the inflow. Self-employedworkers and entrepreneurs, who were expected toestablish and run a business in Canada, had a spe-cial place within the point system from the start. InJanuary 1986, the government added the investorclass, applicants within which gained admission byinvesting specified amounts in Canadian enterprises.While the business classes have never become morethan a small part of the inflow,7 this represents aphilosophical shift toward an idea that immigrationcould be used as a source of capital and as a meansof establishing trade links.8 Notice that these, too,are long-term goals of immigration policy.

At the same time as these changes were beingimplemented, the government undertook a demo-graphic review to examine the consequences of de-clining birth rates and their implications for immi-gration policy. That review, which was presented in1989, indicated that immigration policy was not agood tool for meeting the demographic goals set outin the earlier policy statements. In particular, with afertility rate of 1.7, gross flows that were likely tobe politically feasible could not provide continualpopulation growth. Further, the review showed thateven wild immigration scenarios with large inflowsand 50 percent of the inflow under age 15 wouldnot have a substantial impact on the age structure(Canada 1989). Immigration was not the solutionfor a naturally aging population.

In the early 1990s, the government moved awayfrom demographic goals. At the same time, it beganto increase the importance of the economic compo-nent of the inflow.9 Beginning in May 1991, a des-ignated occupations list was formed which containedoccupations in short supply in specific provinces.Immigrants who matched this list were given addedpoints and processed on a high priority basis. Mostimportantly, in 1992 the Conservatives introduceda new Immigration Act which was designed to pro-vide greater control over the inflow. The Act gavethe department broad new regulatory powers, includ-ing the power to set limits on components of theinflow and turn away applicants once specific num-bers had been reached for a particular category. This,potentially, was a move toward increased use ofimmigration for economic policy. Indeed, the gov-ernment proposed to use the new Act to reduce theproportion of the inflow who were in the family classfrom 52 percent in 1992 to 43 percent by 1995.

Strangely, a new set of regulations proposed alongwith the Act did not make use of these powers. Whilethe regulations proposed setting targets for the vari-ous components of the inflow, the assessed class wasstill a residual. The government also committed tostable inflows of about 1 percent of the current popu-lation and increased the size of the inflow to nearly



250,000 in 1993 in spite of a persistently poor la-bour market. This was the first time that a Canadiangovernment had responded to high unemploymentrates by even maintaining, let alone increasing, thesize of the inflow.

RECENT POLICY

Recent immigration policy can be seen as an exten-sion of policies established in the early 1990s. Themost comprehensive blueprint for the Liberals’ strat-egy is found in the 1995 document, Into the 21stCentury: A Strategy for Immigration and Citizen-ship. The key elements of the new framework are asfollows. Immigration levels are to be maintained at1 percent of the population level, though targetranges rather than specific numbers are to be set.Within that rough total, refugee management is tobe moved into a separate system with separate re-sources and goals from the rest of the inflow. Thereis a clear commitment to achieving approximateequality of the family class and assessed inflows,even if that means falling short of overall level tar-gets. For instance, Into the 21st Century proposesplacing family-class applicants other than spousesand dependent children in a separate class andputting numerical limits on that class. All of theseelements point to a continuation of the trend towardan enhanced role for the economic component ofthe inflow. Indeed, it is the Liberal government thathas met the goal stated by the Conservatives of hav-ing the family component make up no more than 45percent of the inflow.

What is the goal of this enhanced economic com-ponent? Statements in Into the 21st Century and aset of proposed regulations introduced in Novem-ber 1995, but subsequently withdrawn, provide someguidance. Those statements indicate that the gov-ernment no longer perceives filling short-term oc-cupational gaps as an important goal. In tellingwording, the government states that filling preciseoccupational niches is “not always effective in meet-ing long term needs” (Canada 1995). The 1995 pro-

posed regulations contained a reworking of the pointsystem with emphasis on broad occupational classesrather than specific categories. However, politicaltrade-offs mean this switch away from targetingoccupations will not be complete. In Into the 21stCentury , the federal government commits to “pro-vide provinces with the opportunity to choose anumber of independent immigrants who meet pro-vincial economic objectives” (Canada 1994b). Im-migrants will be selected to meet specific skill needsin specific regions, as denoted by the provincialgovernments, with applicants in this category givenextra points and high processing priority.

The current goals seem, instead, to be more longterm. Specifically, immigration policy is to be used tochange the nature of the Canadian workforce: “Theproposed changes (in immigration policy) seek to im-prove the skills, flexibility and diversity of the Cana-dian workforce responding to Canada’s new, emerg-ing economy” (ibid.). Notably, the government pro-posed a program be established to identify occupationswhere there is a shortage of labour and which areclosely related to the skills of a specific immigrant.Thus, rather than making admission conditional onthere being a labour shortage in the immigrant’s in-tended occupation, the government would bring inskilled workers and help them to search for a job, pos-sibly in a different but related occupation to their own.

The most telling evidence of a switch towardlong-term goals is the virtual abandonment of ab-sorptive capacity as it had been interpreted by pre-vious governments. This is evident in MinisterSergio Marchi’s statement accompanying the 1994Immigration Plan:

Periodically throughout our history — especiallyduring economic downturns — there have beencalls to slam the door shut to immigration.... Ibelieve that such sentiments indicate lack of vi-sion of what this country can become.... I believethat decisions about immigration should be madefrom the perspective of a long-term vision forCanada’s future (Canada1994a).



The main evidence that the government has aban-doned absorptive capacity is the maintenance oflarge inflows in the face of persistent high unem-ployment. The level of the inflow has been reducedin recent years, but to nowhere near the extent onewould predict based on earlier responses to boutsof high unemployment. To make this point, we re-gressed numbers of landed immigrants in each yearfrom 1946 to 1989 on a constant and the detrendedannual unemployment rate lagged by one year.10

This is meant to show the correlation between im-migration and the business cycle over a period inwhich absorptive capacity appeared to be part ofpolicy decisions. Figure 2 contains a plot of the ac-tual immigration numbers and a fitted line showingpredicted immigration levels based on our simpleregression. Note that the fitted line from 1990 to1995 represents a prediction of the levels that wouldhave been in place over this period if absorptivecapacity were still important. The fitted line tracksactual immigration levels quite well before 1990

with the exception of the spike created by the 1956Hungarian refugee movement. In contrast, there isa large gap between the actual and predicted inflowsin the 1990s; while the prediction based on extend-ing the pre-1990 absorptive capacity policy turnsdown in the 1990s because of rising unemploymentrates, the actual levels continued to rise. By 1993the gap between the actual and predicted levels was150,000. Even with recent reductions in the immi-gration level, the difference is approximately 70,000.The gap is a measure of the extent to which bothgovernments in this period have moved away fromearlier policies. The abandonment of absorptive ca-pacity signals a victory of the long-term over theshort-term view of the benefits of immigration. In-stitutionally, the recreation of a separate Departmentof Citizenship and Immigration in 1993 can be seeneither as the source or a signal of the shift in power.

Several facets of the new immigration programare tailored to the goal of, “ensuring that newcomers

FIGURE 2Actual and Predicted Immigration Rates, Canada 1947-1997

27

50 55

2625

121110

Imm

igra

tion

in T

en T

hous

ands

Year

60 65 70 75 80 85 90 95 10001234

9

45

Actual ImmigrationPrediction Based on Unemployment Rate

5678

131415161718192021222324



to Canada can integrate and contribute to Canada asquickly as possible, without adding to the burdenon social programs” (Canada 1996). This is thesource of new policy proposals which would makeimmigrants pay for settlement services and makesponsors post bonds out of which any welfare useby the sponsored immigrant in the first ten yearsafter arrival would be subtracted. It is also a sourceof the push for more flexible workers since, in the“new” economy where constant change is to be ex-pected, more flexibility will mean less time collect-ing social security. Most important in this regard isthe increased emphasis on language proficiency inthe selection process. Immigrants fluent in Englishor French are argued to enter the labour market morequickly. Finally, the government has maintained acommitment to using immigrants to generate invest-ment and trade flows.

SUMMARY OF MAIN TRENDS

The policy periods we have examined can be aggre-gated into five broad periods. The first covers thetime before approximately 1960. This is a periodwhen short- and long-term economic goals for im-migration policy were in concert. The governmentwanted both more people in general to populate anempty land and more unskilled people in particularto provide the labour to develop our natural resourcebase. The specific goals changed over time fromunskilled labour generally before World War I tofarm labour for the west in the 1920s to unskilledlabour for mining and forestry in the 1950s. Over-laid on top of this was an attempt to tailor the sizeof the inflow to the absorptive capacity of theeconomy. This meant virtually shutting down im-migration during economic downturns. From 1960to 1978 the economic goal of immigration policychanged to the short-term one of matching skill gapsin the occupational structure. The point system waspart of an attempt to do that matching as were at-tempts to keep the size of the sponsored componentof the inflow small. From 1978 to 1986, the eco-nomic goals of immigration policy were swamped

by humanitarian goals. The government maintaineda commitment to targeting immigrants to skill gapsbut their ability to do so was hampered by a systemin which applicants assessed for their skills were aresidual category relative to refugees and familyclass immigrants. Then from 1986 to 1989, the gov-ernment focused on long-term, demographic goalsfor immigration, placing little emphasis on select-ing specific skills.

Up to the end of the 1980s, Canada’s immigra-tion policy could be broadly summarized as one withlow inflows in economic downturns alternating withlarge inflows directed at specific goals in bettertimes. This time-honoured pattern was abandonedin the 1990s, with the government deciding thatlong-term benefits of immigration were sufficientto justify maintaining large inflows in the face ofhigh domestic unemployment. Further, the long-termgoals being pursued are different from those at anyprevious time: they are not the broad demographicgoals of expanding the population as an end in it-self or of changing the age structure. Instead, thenew long-term goal appears to be to use immigra-tion to change the nature of the Canadian workforce,making it more skilled and flexible. We have neverbefore tried to use immigration policy to change theskill mix in the economy without also stating ex-plicitly which skills are lacking, that is, without try-ing to micro-manage the process.

WHAT SHOULD THE ECONOMIC GOALS OF

IMMIGRATION POLICY BE?

We turn now to assessing immigration as an eco-nomic policy tool for Canada. Rather than directlyevaluating current immigration policy, we pursue themore general end of examining a set of potentialeconomic goals for immigration. Once that evalua-tion is complete, an understanding of the subset ofgoals characterizing current policy will follow. Thespecific set of goals we examine are those that havebeen pursued in Canada’s immigration policy at leastonce in our history. Focusing on this set allows us



to narrow our discussion to the goals that have beendeemed relevant for Canada. Our approach in con-sidering each goal is to ask whether immigration isthe best policy for meeting that goal rather than dis-cussing the usefulness of the goal itself. Thus, wewill consider immigration as one of a group of (pos-sibly) competing policies. We argue that, unlike inthe past when there were identifiable goals for im-migration policy that could not be met through al-ternative policies, there is currently no single goalfor immigration policy that could not be met betterthrough an alternative.

Before evaluating any policy, we need to decidewho we care about in the evaluation. A commonapproach is to examine the impact of immigrationon individuals resident in the host country beforeimmigration occurs. This, however, leads to poten-tial inconsistencies as the optimal policy may changeonce the new immigrants arrive in Canada and theirutilities are included in any calculations.11 As anextreme response to this consistency problem, wecould take account of the utilities not only of resi-dents and immigrants but also of potential immi-grants. This approach also has difficulties, whichare evident in the context of a classical utilitariansocial welfare function. Working with such a func-tion, we could end up at the conclusion that weshould bring in a large number of immigrants evenif doing so meant very low average utilities foreveryone in the country. The latter policy essentiallytrades off the standard of living for sheer numbersof individuals and arises because the average utilitylevel and the number of people in a population areperfect substitutes in a classical utilitarian socialwelfare function. In response to these problems,Blackorby and Donaldson (1984) introduce an ap-proach called critical level utilitarianism (CLU).Under CLU, one evaluates different social states byadding up the lifetime utilities of all residents andpotential residents after subtracting from each afixed, critical level of utility, α. This critical levelacts like a fixed cost of adding extra people so thata higher level of α implies that society places moreemphasis on individual levels of utility and less on

having more people. Thus, we can analyze variousimmigration policies not just by asking whether itincreases total or average utility but by askingwhether it results in more people (including theimmigrants who would be brought in under thepolicy) having more than a minimum standard ofliving. It is this CLU approach which we will im-plicitly consider in the discussion that follows.

Our next step is to sketch out an economic modelas a framework for discussion. The model we willconsider is a standard one presented in Borjas (1995)among other places. In this framework we assume aproduction function f (K,L) which is a function ofthe capital, K, and labour, L, used in the economy.For the moment, immigration will be assumed tocause an increase in L but have no effect on K. Per-haps the most crucial assumption at this point isabout the returns to scale embodied in the produc-tion function. Trefler (1997) shows that if the func-tion embodies a constant returns-to-scale technol-ogy then increasing L without changing K may meanthat average income will stay constant or fall, de-pending on assumptions relating to trade and factormobility. If, in contrast, the function embodies anincreasing returns-to-scale technology then immi-gration can lead to growth both in total output andin income per capita (i.e., to both extensive and in-tensive growth). Intensive growth could occur forseveral reasons. First, more immigration increasesthe size of the market, implying more output and,perhaps if there is “learning by doing,” a resultantincrease in productivity. Second, Simon (1989) ar-gues that more labour means more potential inno-vators, which in turn means productivity-enhancingimprovements in technology. Third, there could befixed costs to production implying that the greaterproduction accompanying a larger economy willbring efficiencies.

Is there any evidence that there are economies ofscale related to population growth in moderneconomies? Cross-country examinations of the ef-fect of population growth on per capita income gen-erally find no such effect. For example, Barro and



Sala-i-Martin (1992), using Japanese and US data,find that net migration to a region has almost noimpact on economic growth for that region in esti-mation where they attempt to address the simulta-neity between growth and migration flows. EvenSimon, a strong proponent of the position that popu-lation growth promotes economic growth, admits“empirical studies do not prove that fast populationgrowth in the more developed world as a whole in-creases per-person income”(Simon 1989, p. 197).Of course, as Simon points out, the opposite is alsonot true: population growth in modern economiesdoes not appear to be related to reduced economicgrowth as would be predicted with a constant re-turns-to-scale production function and a fixed capi-tal stock. Thus, a simple alternative would be thatcountries face a constant returns-to-scale technol-ogy combined with a perfectly elastic supply of capi-tal. In this case, immigration could expand the sizeof the economy but could not lead to intensivegrowth.

We next need to make assumptions about the na-ture of trade for the economy since trade and mi-gration are potentially substitutes in terms of theireffects on returns to factors in Canada. Given theexistence of the North American Free Trade Agree-ment (NAFTA) and movements toward trade liber-alization under the General Agreement on Tariffsand Trade (GATT), it seems reasonable to us to startwith the assumption that there is free trade betweenCanada and other countries, that is, that Canada isan open economy facing few legislated barriers totrade.

The composition of the inflow will also be im-portant in determining its impacts on the hosteconomy. Thus, we will split labour into two com-ponents: unskilled labour, Lu, and skilled labour,Ls. If we maintain the constant returns-to-scale as-sumption, assume that labour immigrates in the sameratio of skilled to unskilled as exists in the economyand assume a perfectly elastic supply of capital thenthe conclusions on immigration and economicgrowth just discussed do not change. Furthermore,

relative factor prices do not change and so distribu-tion is not affected. If there is a disproportionatenumber of immigrants of one skill type or another,however, the outcomes wil l depend on thesubstitutability and complementarity between fac-tors. For example, if skilled and unskilled workersare complements and the immigration inflow con-tains only skilled workers then wages for the un-skilled will rise and those for the skilled will fall.Thus, the composition of the inflow in this modelcan affect distribution among owners of differentfactors in the economy. Given the CLU frameworkmentioned above, we are probably more concernedwith impacts on the unskilled than on other groups.

Based on these discussions, we will take a modelwith constant returns-to-scale technology, three in-puts (capital, unskilled labour, and skilled labour),a perfectly elastic supply of capital and free tradeas our starting point for discussing various specificpotential benefits from immigration for Canada. Thisis clearly a model biased against a conclusion thatimmigration has a significant positive impact onintensive economic growth. As we focus on eachpotential economic goal for immigration in turn, wewill consider additions to the model that implygreater benefits from immigration. Of course, theseadditions may imply advantages to other policies aswell, and our focus will be on whether immigrationis the best policy to reap any implied benefits.

Filling Occupational “Gaps”One main way in which immigration is argued tobenefit a host economy is through filling “gaps” inthe occupational structure. Filling such gaps couldbe beneficial if the production function is of a formsuch that minimal amounts of specific inputs arerequired or if there are empty sectors in the economywith increasing returns to scale when they are firstdeveloped. In the extreme, one could imagine aLeontief production function where inputs must beused in fixed proportions. In that case, even if theother factors are growing, we may not be able toachieve either intensive or extensive economicgrowth without sufficient labour. This type of



rationale was behind the earlier immigration poli-cies designed to “fill the west” and the targeting ofskilled workers after 1960.

Whether such extreme complementarities existin current production functions is unclear. However,our goal is not to decide on the exact structure ofproduction but to discuss the usefulness of immi-gration policy in meeting stated targets. Thus, as-sume that some key occupations exist and that weneed to fill them. Is immigration policy the best tool?In the period before 1960, what was needed was alarge and, given the race in the early part of the pe-riod to fill the west before the US took control of it,rapid increase in population. There does not appearto be an alternative policy to immigration that couldhave met these goals; changing fertility rates wouldnot have provided a fast enough response. Green andSparks (1999) show that immigration was a substan-tial contributor to both intensive and extensive eco-nomic growth in Canada before World War II.

Arguments about filling gaps at the more skilledend of the occupational structure are less clear sincefeasible, alternative policy responses do exist. Inparticular, if there is a skill shortage, the govern-ment could respond through training and educationfor individuals already resident in Canada. The needin this case is not for more people but more skills.At the time when the policy decision to shift towarda more skilled workforce was made (i.e., around1960), the educational infrastructure did not existto upgrade the skill level of the Canadian workforce.In this circumstance, immigration was the only fea-sible tool for rapidly changing the skill composi-tion of the workforce. Moreover, some skilled im-migration was necessary to provide the training forthe native born. Thus, the government used the poli-cies of bringing in skilled immigrants and expand-ing the educational system as complements. Oncethat system was expanded, however, increasing skillsthrough immigration and through education of theresident population became substitutes as policies.12

Recent large increases in enrolment among youthssuggest that the education system can be used to

meet increased skil l requirements (Card andLemieux 1996). The question then becomes whetherwe want to use education or immigration policy tomeet specific skill requirements.

Exactly how substitutable immigration and edu-cation policies are depends in part on the assumedmodel of human capital generation. In the simplesthuman capital model, in which everyone has equalability and free access to the capital market to fi-nance education, the wage differential betweenskilled and unskilled workers will be just enough tocompensate skilled workers for their skill invest-ment. In that case, bringing in skilled immigrantswill not change the eventual distribution of skills inthe economy or the distribution of wages betweenskilled and unskilled workers. Thus, education andimmigration skill policies would be perfect substi-tutes. If, instead, there is a wide range of abilitiesand imperfect access to capital markets to fund edu-cation then the wage differential between skilled andunskilled workers may more than compensate forthe educational investment of skilled workers.Moreover, bringing in a certain number of skilledimmigrants will not necessarily lead to an equalnumber reduction in native-born individuals whoacquire skills. Assuming demand for skilled work-ers is not perfectly elastic, it will still lead to a re-duction in the skilled/unskilled wage gap. It seemsreasonable to assume that there are some individu-als who are just on the margin of deciding to go touniversity given any observed wage gap and, con-sequently, that a reduction in the gap will cause somereduction in the number of native born who acquireskills. Thus, it is reasonable to assume that educa-tion policy and a skill-gap filling immigration policyare substitutes to some degree.13

If education and immigration policy are substi-tutes, which should we use to fill occupational gaps?One apparent advantage of using immigration policyis that Canada does not have to pay for the humancapital investment in the immigrants; that is donein the source country. The main argument againsttaking advantage of such a benefit must be moral:



purposefully attempting to confiscate the humancapital investments of other countries, particularlyless developed countries, should not be a goal ofour immigration policy.

A second potential advantage to filling skill gapsthrough immigration policy is that more direct tar-geting appears to be possible. With immigrationpolicy, the government sets admission screens to letin the desired occupational group. With educationalpolicy, the government can only set up incentivesand hope that individuals respond. Moreover, it ap-pears that we should be able to fill any gaps morequickly using immigration rather than domestic skillgeneration. These arguments have several difficul-ties, however. First, immigrants may have difficultygaining acceptance of their educational credentialsand thus difficulty in getting a job in the occupationfor which they were admitted. A common findingthat there is a lower rate of return to education forimmigrants relative to non-immigrants (Abbott andBeach 1993; Baker and Benjamin 1994) points tosuch difficulties. Similarly, using data from a panelof immigrants from the early 1970s, Green (1995b)finds that obtaining credentials recognition is a sig-nificant contributing factor in explaining rapid earn-ings growth among immigrants in their first threeyears after arrival. A second, related difficulty is thatimmigrants often take a long time to enter employ-ment. Thus, Green (1999) and Green (1995b) showthat immigrants experience very high unemploymentrates of over 20 percent in the first six months afterarrival. These decline rapidly with time in theeconomy, but they still indicate that immigrantsshould not be assumed to be an instant source ofskill supply. In comparison, training residents insome trades takes less than two years and thus mightprovide comparable response times to immigration.For other occupations the training time will belonger than the immigration response. Finally, im-migrants may not move into the occupation to whichthey are targeted. Green (1995a) compares the dis-tribution of stated Intended Occupations (IOs) ofimmigrants when they arrive to the actual occupa-tional distributions of samples of immigrants from

the same arrival cohorts in subsequent Censuses andfinds that “leakage” into occupations other than theIO is not trivial.14

Perhaps the most important argument against try-ing to use immigration to fill specific skill gaps isthat it is too difficult to accurately identify and tar-get flows to those gaps. Green and Green (1995)find that the point system can work in increasingthe numbers of immigrants intending to enter speci-fied occupations but that with so many other goalsfor immigration policy as a whole and the point sys-tem more specifically, successful targeting has be-come rarer since the early 1970s. Thus, even if thedifficulties in immigrant adaptation to the neweconomy set out above do not exist, we would needto make occupational targets one of a small set ofgoals of immigration policy if they are to work atall. But this too will be of little use if we cannotaccurately identify the skill gaps we are trying tofill. In a review of related evidence, the EconomicCouncil of Canada concluded that “there is no reli-able way to detect gaps in the labour market” (Eco-nomic Council of Canada 1991, p. 9). As seen inthe second section, the government itself appears tohave lost faith that it can identify those gaps. Thisis the reason for its proposal to target a broader rangeof skills and to help immigrants find gaps after ar-rival. But if the workers themselves are to play alarge part in identifying gaps then the evidence thatimmigrants take time to adapt to their new economysuggests that residents would likely perform this taskbetter.

Finally, it is worth reiterating that there is no firmempirical evidence that population growth leads toeconomic growth among advanced countries. If fill-ing skill gaps were necessary for Canada to succeedin the race to develop and implement new technolo-gies then one would expect that we, as one of thefew countries to try to use immigration policy tofill such gaps, should have better economic growththan our developed partners over the last 25 years.This does not appear to be the case.



Expanding Human CapitalWe turn next to the possibility of using immigra-tion to increase the stock of human capital. To aidin our discussion, we can alter our model such thatthe production function includes human capital as aseparate input. One could then consider whetherexpanding the stock of human capital can lead tointensive as well as extensive economic growth. Theargument that more human capital could generateincreasing returns to scale, perhaps through increas-ing innovation, seems plausible. However, if thiswere true then countries with greater stocks of hu-man capital should experience greater growth in in-come per capita. A comparison of the intensivegrowth over the last quarter century in Canada andthe United States, which have similar natural en-dowments but are very different in the sizes of theirrespective human capital stocks, suggests this is nottrue. Thus, we are again led to the conclusion thatexpansion of any type of labour without bound isnot called for.

Whether or not we are seeking unlimited quanti-ties of human capital, but particularly if we are not,the issue of a trade-off between education and im-migration policy must again be addressed. Both theeducation system and immigration can be used toincrease the human capital stock. As was the casewith filling specific skill gaps, using immigrationpolicy to import human capital will decrease wagedifferentials between more and less skilled workersand thus reduce the human capital acquired by resi-dents. Thus, immigration and education policy aresubstitutes. Both seem capable of generating sub-stantial increases in human capital. Coulson andDevoretz (1993) show that the value of human capi-tal transferred to Canada through immigration overthe last 20 years has been substantial and could havebeen larger with more careful management. At thesame time, we know that the proportion of the Ca-nadian population with a university degree is muchsmaller than the equivalent proportion in the UnitedStates (0.16 vs. 0.26 in 1989), indicating that thereis also room for expansion of the capital stockthrough education in Canada. The fact that enrol-

ment rates have sky-rocketed in the 1990s in Canadaindicates that both the infrastructure and the desireexist to realize this potential.

Which policy should we use? As in the discus-sion of filling skill gaps, any arguments in favourof using immigration to increase the skill level arereduced to the extent that immigrants have difficultyin gaining recognition for their credentials. The mainargument in favour of using immigration to increasehuman capital is that Canada does not have to payfor the investment. As we argued before, however,robbing other countries of their human capital in-vestments does not appear to us to be a morally de-fensible goal for our immigration policy. Further,there appear to be reasons stemming from the cur-rent state of the Canadian labour market that pointto using domestic education policy. We know thatthere have been increasing university/high schooldifferentials in earnings and employment rates foryoung men and women but sharply declining earn-ings across successive labour market cohorts of men(Bar-Or et al. 1995; Beaudry and Green 1997). Onepossible policy response to this growing inequalityis to help more young people acquire the highereducation and the higher earnings that go with it.Should we seek to expand the supply of the univer-sity educated from another source when the earn-ings of new, university educated males are fallingand just when the desired expansion of human capi-tal appears to be happening on its own? While wemay want a very skilled workforce in terms of thedistribution of education among its members in or-der to manipulate new technologies, there does notappear to be a clear argument that the optimal stockof human capital is bigger than what could be cre-ated through educating the resident population.Without such a clear argument, it seems difficult tojustify increasing the human capital stock as a maingoal of immigration policy.

Meeting Regional NeedsAnother proposed goal of immigration policy is totarget labour and skills to regions of the country withsmall endowments of either or both. The argument



in favour of using immigration in this way must bethat the regional economies need certain sets of skillsin order to grow. Again we raise the question, if thisis the problem, is immigration the best policy solu-tion? Here the main difficulty with using immigra-tion is that immigrants might not remain in needyareas even if their admission were based specificallyon their doing so. Once immigrants enter they shouldbe and are entitled to the same rights as other resi-dents, including the right of free movement withinCanada. Dougherty (1994) follows the migrationpatterns within Canada of immigrants using admin-istrative data and finds that between 15 and 20 per-cent change provinces within six years after arrivalin Canada. Further, immigrants seem quite respon-sive to income levels and unemployment rates inchoosing their destination. Lin (1997) compares in-terprovincial mobility of immigrants and non-immigrants using data from the Labour Market Ac-tivity Survey and finds that their migration rates ina given year are not statistically significantly dif-ferent from one another. The suggestion from thesestudies is that immigrants are mobile, though notinordinately so. The difficulty with this evidence isthat it does not answer the main question of inter-est: if we forced new immigrants to go to depressedregions of the country, perhaps by making initialresidence in a province a condition of their admis-sion, would they stay? The Dougherty and Lin pa-pers examine the 1980s, a time when immigrantswere free to choose their initial location. TheDougherty finding of substantial responsiveness ofimmigrants to regional income and unemploymentdifferentials suggests that they might be highlymobile if someone else chose an initial (high unem-ployment) location for them. Indeed, the fact thatimmigrants choose disproportionately not to settlein the Atlantic provinces suggests that they couldnot be successfully used as a long-term source oflabour for that region. One recent attempt to forceimmigrants to match particular provinces is in theimmigrant investor program. Immigrant investorsenter by investing funds in a particular province butdo not have to reside in that province. As a result,while British Columbia received 11.8 percent of all

investment under this program in 1993, it was theresidential choice of 54.1 percent of the investors(Kunin and Jones 1995). Thus, the program steeredfunds but not people to the slower growingprovinces.

Generating InvestmentOne main component of the current immigrationinflow, the immigrant investor program, concen-trates on using immigration to generate investmentin Canada. Applicants can enter in the investor pro-gram if they have a minimum net worth and investat least a specified minimum in small or medium-sized enterprises in Canada. The investments mustfall in a set of pre-approved categories and must bedeemed to be likely to generate employment forCanadians. In an economic model in which the hostcountry is assumed to have a fixed capital stock inthe absence of this program and in which there is aconstant returns-to-scale technology, immigrationwill lead to a decline in wages as more labour isapplied to a fixed capital stock. A program thatbrings in capital to match the immigrant labour in-flow could reverse the predicted wage decline, help-ing the low-skilled labourers about whom we aremost concerned.

The key question in determining the usefulnessof an immigrant investor program is Canada’s open-ness to international capital flows. In the model de-scribed earlier, in which there is a perfectly elasticsupply of capital, an immigrant investor programwill have no net impact. Any funds brought inthrough this route would depress rates of return inCanada below the going world rate and lead to anexodus of other funds until the rate of return risesback to the world rate. Previous evaluations of theimmigrant investor program of which we are aware,such as Kunin and Jones (1995), simply add up thedollars officially invested and the jobs associatedwith those investments to calculate the impact ofthe program on the Canadian economy. This wouldonly be an accurate expression of the net impact ifthere were no capital mobility and the projects be-ing invested in would otherwise go unfunded. At this



point, we do not know whether bringing in immi-grant investors has positive net benefits but, giventhat world capital markets are quite open, it seemslikely that the net benefit is much smaller than thegross investment dollar total would indicate. Tomake generating investment a main goal of immi-gration policy would seem unwise in the absence ofmore substantial evidence of net benefits from sucha program.

Increasing Trade FlowsA more recently stated goal of immigration policyis to help increase trade flows. The idea here is thattrade connections depend in part on trust and infor-mation. Potential exporters and importers need tohave information that it is viable in order to enter aparticular market in another country and need toconnect with suppliers and distributors they cantrust. Cultural and legal differences across countriesmay make it difficult to establish the types of busi-ness links needed to carry on the mechanics of trade.Immigrants from a particular country would presum-ably have connections with specific potential sup-pliers and distributors in their country of origin, haveinformation about the state of markets in that coun-try and have the kind of cultural and institutionalknowledge about both their country of origin andCanada that would facilitate trade between the two.Globerman (1995) proposes this as a potential ben-efit from immigration but, based on basic tabula-tions, casts doubt on whether any such benefit islarge. Head and Ries (1998) estimate the impacts ofthe stock of immigrants from a given source coun-try in Canada in 1991 on the volume of trade be-tween Canada and that country. They find that in-creasing the immigrant stock from a given countryby 10 percent increases exports to that country by1.5 percent and imports from that country by 3.8percent. Thus, immigration does affect trade, thoughmore by affecting imports than exports. Interest-ingly, Head and Ries find that the best predictor forincreased trade volumes with a given source coun-try is not the number of entrepreneur and investorclass immigrants from that country but the numberof other independents.15 This indicates that there is

no reason to focus on business class immigrants ifwe choose to use immigration policy for promotingtrade.

Again, the key question is whether immigrationis the right policy tool for increasing trade. Theproblem being addressed in this case is one of in-complete markets in information. This problemcould be solved with a small set of firms selling in-formation about overseas markets to Canadian ex-porters. Beyond some minimal number of suchfirms, bringing in more immigrants from a particu-lar source country will yield diminishing addedamounts of information about that country. Thus,added immigrants would serve to dissipate rentsbeing earned by the first firms that sold the infor-mation rather than bringing large net benefits toCanada. That there might be some such non-linearityin the relationship between trade benefits and mi-gration flows is indicated by the fact that Helliwell(1997) finds no impact of stocks of immigrants fromother provinces on interprovincial trade in Canada.This could arise if information enabling trade amongprovinces is so widespread that there is no addedtrade benefit to extra migration. Thus, while Headand Ries’ results may serve as a justification forsome immigration they are not necessarily justifi-cation for large, ongoing immigration levels.

Altering the Age Structure of the PopulationImmigration could conceivably be used to offsetcurrent tendencies toward the aging of the residentpopulation. With the aging of the baby boomers,there is considerable concern about a future rise inthe ratio of dependents (non-workers) to workers inthe Canadian economy and the implication of thatrise for public pension provision. One could imag-ine offsetting this aging tendency by bringing inlarge numbers of young immigrants. However,Denton, Feaver and Spencer (1997) argue that, giventhat a normal immigration inflow includes its owndependents, even very high immigration would haveonly marginal impacts on the age structure. Thus,they project that with annual inflows of 200,000immigrants per year, the ratio of those out of the



labour force to those in the labour force will be 24.8percent in 2036 (approximately twice what it is now)while with annual inflows of 500,000 that depend-ency rate would be reduced only slightly to 21.3percent. This supports the results from the 1989Demographic Review. The review calculated thatwith inflows of 200,000 immigrants per year, ap-proximately 22 percent of the population would beover age 65 in 2036 compared with approximately17 percent if we brought in 600,000 immigrants peryear with 50 percent of them being under age 15.Thus, even a policy of bringing in large numbers oforphans cannot solve problems related to the agingof the population.

Increasing FlexibilityImmigrants are sometimes assumed to supply a spe-cial type of labour, superior to domestic labour sup-ply in its flexibility. Given that immigrants are bydefinition mobile and that they may have fewer so-cial ties to lock them into particular communitiesor jobs in Canada, it is reasonable to predict thatthey would be more mobile than native-born work-ers. If so, then immigrants could provide the greasethe economy needs to adjust to various shocks: im-migrants would be the workers who respond mostto wage signals, equalizing the marginal product oflabour across various regions and occupations.Green (1999) argues that immigrants are in fact moreoccupationally mobile than native-born workers and,specifically, that they moved out of the decliningmanufacturing sector more rapidly than native-bornworkers in the 1980s. On the other hand, the regionalmobility evidence cited above does not indicate thatimmigrants are more geographically mobile than thenative born. Mobility can be viewed as a form of in-vestment like human capital investment and thus, someof the same arguments raised in discussing human capi-tal stocks are relevant here. As in that case, bringingin more mobile immigrants will dampen wage signalswhich might otherwise induce native-born workers toleave depressed areas. However, domestic policy aimedat inducing the native born to make particular occupa-tional or regional adjustments seems as unlikely tosucceed as trying to target immigrants at particular

occupational gaps. Thus, immigration seems to be asuperior policy in that it simply involves bringing inmore mobile individuals and letting them select thebest response to economic shocks. How many immi-grants we would want to bring in to fill this sort of“shock troop” role is unknown.

Increasing Returns to Scale ArgumentsAs discussed earlier, based on existing econometricevidence, we believe that a constant returns-to-scaletechnology best describes the situation facing de-veloped economies. There are no gains in terms ofintensive economic growth to be had simply fromincreasing the population size. In the spirit of theremainder of the discussion, however, suppose thiswas not true. Does this imply that Canada shouldseek large immigration flows? If a larger scale ofproduction means more productivity then it seemslikely that a small country such as Canada couldachieve much greater efficiencies through trade thanpopulation growth. This raises the whole issue ofinteraction of trade and labour flows.

In a standard Hecksher-Ohlin model with twogoods, two countries and two factors of production,trade patterns are based on the relative factor en-dowments of the two countries. If both countriesproduce both goods in the trading equilibrium thenthe factor prices are equalized between the coun-tries even without migration. Permitting migrationdoes not change anything: factor flows and tradeflows are substitutes. However, if the countries spe-cialize in production of one of the goods then factorflows can lead to greater world efficiency. If capitalis perfectly mobile, however, wage equalization willagain follow without any migration. Thus, in a worldwhere both trade and capital appear to be more mo-bile than labour, it is not clear that there are largeextra gains from migration.

In an alternative model, trade is based on econo-mies of scale in particular industries rather than rela-tive factor endowments. In such a world, one mightconsider using immigration to bolster the size of theCanadian market to take advantage of those scale



economies. With relatively free trade with the UnitedStates, however, increasing our population increasesthe market not only for our own firms but also forUS firms. If there is an increasing returns-to-scaletechnology industry in which the US already pro-duces more (and hence is more efficient), an increasein our population may not allow Canada to expandits production in that market. Whether our industrywould expand is unclear and depends on factor pricereadjustments. The point, though, is that for a smallopen economy such as Canada, immigration is nota clear path to taking advantage of scale economies.Similarly, arguments about immigration increasingproductivity through innovation do not necessarilyfollow even if one believed that more populationmeant more innovation. Whether Canada would ben-efit more from an innovator being in Canada ratherthan their original country is not clear, particularlyif Canada is not willing to apply more resources toresearch than the potential source country.

It is worth noting that, while there does not ap-pear to be evidence of particularly strong positiveimpacts of immigration on intensive growth, therealso does not appear to be evidence of strong nega-tive impacts. Examining the mainly US literatureon the impact of immigration on the native born,Friedberg and Hunt conclude, “empirical estimatesin a variety of settings and using a variety of ap-proaches have shown that the effect of immigrationon the labour market outcomes of natives is small”(1995, p.42). Similar evidence for Canada does notexist and, given that we have only a small amountof variation in local labour markets to identify suchan impact, may never exist in a convincing form.The differences in our immigration policy relativeto the US indicate that conclusions for the US can-not be easily imported but at least the evidence fora similar economy does not imply large negativeimpacts. Pope and Withers (1993) use macro datato examine immigration impacts for Australia, whichhas a similar immigration policy to Canada’s. Theyconclude that immigration can actually lead to de-clines in unemployment though any such declinesare at best small. Finally, using cross-national data,

the Economic Council of Canada finds evidence for“very small” positive impacts of population growthon intensive economic growth and no effect of popu-lation growth on the unemployment rate (EconomicCouncil of Canada 1991). Earlier we used this mini-mal impact result to argue in favour of an economicmodel with constant returns-to-scale technology anda perfectly elastic supply of capital. An alternativemodel that does not require perfectly elastic capitalsupply is one in which immigrants bring some ben-efits (perhaps through impacts on trade linkages andlabour force flexibility) that offset costs implicitwith a constant returns-to-scale technology. Thereis no clear evidence that either raising or loweringthe immigration level would lead any benefits ofimmigration to exceed the costs.

MANAGING THE COMPOSITION OF

IMMIGRATION

The discussion to this point does not indicate clearbenefits from any particular goal of immigration.That does not mean that we should abandon man-aging the inflow, however. Seeking to select immi-grants who will adapt easily to the Canadianeconomy potentially reduces the fiscal costs of im-migration. Thus, it is worth noting that Green (1999)finds that individuals not in the independent classand/or with low language skills at the time of ar-rival are much less occupationally mobile than thoseentering as independents. Further, we may wish tominimize direct impacts on particular groups ofworkers. In this regard, while we have argued againsttrying to steer immigration toward specific predictedskill gaps, it is still reasonable to minimize immi-gration into high unemployment occupations. Fur-ther, even if we did not want to make building hu-man capital stocks a primary goal of immigrationpolicy, it would be reasonable to select immigrantsbased on their education level. Unemployment ratesfor university-educated individuals in Canada in1996 were 5 percent compared to rates of 8 percentfor individuals with a postsecondary education and11 percent for individuals with a high school



education.16 Based on the 1993 Survey of ConsumerFinances, corresponding unemployment rates forimmigrants entering before 1986 were 5.8 percentfor the university educated, 8.5 percent for thosewith a postsecondary certificate or diploma below aBA and 10 percent for those with a high school edu-cation. This suggests that university-educatedimmigrants will be more likely to enter the labourforce successfully. Similarly, in a well-managedimmigration system, bringing in younger applicantsin order not to contribute to the immediate increasein the dependency ratio seems reasonable. Further,an inflow balanced across source countries appearsbeneficial if for no other reason than it will helpencourage a balanced trade pattern for Canada.

The current immigrant selection system appearsto permit sensible management of the inflow in gen-eral. It bars applicants whose intended occupationmatches a list of high unemployment occupations,and attempts to select individuals from around theworld who are young, educated, and fluent in Eng-lish or French. Unfortunately, the system still triesto direct the inflow toward specific occupations boththrough points given for being in a list of accept-able occupations and, more importantly, through theadded points and high processing priority given toimmigrants in occupations specified by the provin-cial governments as being in high demand. The fed-eral government appears to be moving away fromsuch targeting, as we argue is reasonable. Why thenshould we reverse that policy direction by givingthe provinces the power to target immigration atspecific gaps? In standard arguments about how todivide powers in a federation, activities that spillover provincial boundaries should be allocated tothe federal government. The results on regionalmobility discussed above suggest that immigrantsare mobile and that they might be even more so ifwe attempt to coerce them into locations and occu-pations they might not otherwise choose. Based onthis argument, immigration should be a strictly fed-eral power. This conclusion could be reversed to theextent that provincial governments have better in-formation about gaps in their local occupational

structures. There is no reason to believe this is true.The federal government has a system to project oc-cupational gaps but, as we have seen, CIC no longerappears to believe it accurately identifies skill gaps.Why should provincial governments, with fewerresources dedicated to labour research, do a betterjob of predicting such gaps? The British Columbiagovernment’s Work Futures program, which at-tempts to predict occupational openings, is in factbased on the federal government’s occupational pro-jection system (British Columbia 1996). Thus, it ishard to argue that it could be doing a better job ofprediction than the federal government. If there isno clear reason to have the federal government tryto target skill gaps, there is even less reason to havethe provincial governments do so.

CONCLUSION

Our goal in this paper has been to use a combina-tion of Canada’s immigration policy history andcurrent immigration research to examine economicgoals for immigration policy in Canada. We presentthree main arguments. First, Canada’s immigrationpolicy up to the late 1980s was characterized by al-ternating periods of expansion targeted at specificeconomic goals and periods of virtual shut down inresponse to poor domestic labour market conditions.The specific goals that have been targeted have var-ied from long-term goals of increasing populationand altering the age structure to the short-term goalof filling specific perceived gaps in the occupationalstructure. Moreover, in each case one can make anargument for why immigration was the best policyfor meeting the specific goal. Second, the policy inthe 1990s has broken with the established pattern innot cutting back immigration in response to highunemployment rates and in switching almost com-pletely to long-term goals. In this regard, claims thatkeeping the inflow at or near 1 percent of Canada’spopulation is consistent with our policy history aresimply not true. Third, in an examination of a list ofpotential goals, we have not been able to find a con-vincing argument that immigration is the appropriate



policy tool for meeting any specific goal. There issome support for using immigration as a tool of tradepolicy and for increasing the flexibility of the la-bour force, but neither of these goals indicate a needfor large inflows in poor labour market periods.Arguments can be made for using immigration tomeet other economic goals, such as increasing theskill level of Canada’s labour force, but in each casewe argue that other policies would perform better.Thus, Canada appears to have entered a new era inwhich there is no clear economic goal for whichimmigration is best suited.

Does this mean that Canada should abandon im-migration? By no means. A defining feature of im-migration is that it constantly rejuvenates the popu-lation: bringing in people with new perspectives andcustoms. This makes for a potentially rich culturethat is continually being reinvented. Just as impor-tantly, it necessarily negates attempts to create onesimple definition of what it is to be Canadian. Aglance at the ethnic turmoil that curses some of themost culturally static countries in the world indi-cates the benefit we receive from ongoing immigra-tion. Bringing in a continuous stream of people whocome hoping to find a better place provides the on-going energy needed to make Canada a better place.

As we stated at the outset, our examination iscentred exclusively on economic trends and policygoals, neglecting immigration’s role as a cultural andhumanitarian policy. We chose this approach out ofa need to present a focused argument and becauseof our own limitations in expertise. We believe thatimmigration plays a positive, dynamic role in Cana-da’s cultural development but that to see this clearlyrequires a clear understanding of immigration as aneconomic tool since this is where many discussionsappear to end up. Reading this paper should be com-plemented with reading studies on immigration ascultural policy and with experiencing the wonder-ful diversity of current Canadian society. It shouldalso be recognized that many potential benefits, bothcultural and economic, may be realized in genera-tions that follow the original migrants.

Thus, our conclusion is that immigration policyshould now be seen primarily as a cultural, socialand humanitarian policy rather than an economicone. This does not mean, however, that all economicconcerns should be abandoned. Immigration shouldbe selected to minimize negative impacts on the leastwell off. In this regard, government plans to moveaway from targeting specific occupations and towardselecting broad skills seem appropriate. Also, im-migration should likely be cut back in poor labourforce times rather than being kept at high levels inthe hopes of realizing economic benefits for whichthere is little supporting evidence.

NOTES

The authors would like to thank Paul Beaudry, ScottTaylor and participants at a RIIM seminar for helpfulcomment and discussions. Any errors are our own.

1The fact that public declarations and actual policywere so different was made easier in part by the fact thatimmigration policy was being set through Orders-in-Council rather than through changes in the Act. The lat-ter would have entailed more parliamentary and publicdebate. See Green and Green (1997) for a discussion oflegislative flexibility and its impact on the implementa-tion of immigration policy in Canada.

2The preferred countries included Britain, the UnitedStates, the Irish Free State, Newfoundland, Australia, NewZealand, and South Africa.

3The figure plots the ratio of immigration in a year tothe population at the end of the preceding year. The dataare from Leacy (1983).

4For a discussion of the new perceived needs of theCanadian economy see the Report of the Special Com-mittee of the Senate on Manpower and Employment(Canada 1961).

5In a 1992 policy document the government states, “oursupply of highly-skilled workers is not keeping up withdemand. Even though unemployment is relatively hightoday, hundreds of thousands of jobs go unfilled becausethere are no qualified applicants to fill them.... Immigra-tion can benefit the labour market by helping employersfill skilled jobs for which Canadians are not available.



This alleviates structural ‘bottlenecks’ in the supply ofskilled workers” (Canada 1992).

6An applicant could obtain up to 35 points for the com-bination of occupational demand (the government’s as-sessment of the demand for the individual’s three digitoccupation), specific vocational preparation (the skilllevel of the individual’s occupation) and arranged em-ployment (points awarded if the immigrant had a jobwaiting that could not be filled by a Canadian). Addingin points awarded if the intended destination in Canadahad a labour market shortage, yields a total of 40 pointsfor these categories.

7In 1996, principal applicants in the combined busi-ness classes made up only 3 percent of all immigrants.

8See Borowski and Nash (1994) for a description andevaluation of business class immigration.

9See note 5, the quote is drawn from the policy docu-ment setting out the five-year plan.

10The unemployment variable is created by regress-ing annual unemployment rates on a constant and a trend.The difference between unemployment rates and the es-timated trend is used to predict immigration levels. Thevariable is designed to capture the relation of immigra-tion levels to the business cycle. Immigration level andunemployment rate data are from CANSIM. The regres-sion has 43 observations and an R2 of 0.48.

11An extreme example would be importing slaves: thiscould increase societal welfare if we count only slaveowners (existing residents) but decrease it once we alsocount the utilities of the slaves.

12Note that estimates of the degree of substitutabilityor complementarity between immigrant and non-immigrant skilled labour using historic data (e.g., Akbariand De Voretz 1992) reflect past use of immigration andeducation policies to try to address skill gaps. In con-trast, the discussion here centres on filling a specific pro-jected set of job openings. In that context, there is noreason to believe that immigrant and non-immigrants withidentical qualifications would be anything other than sub-stitutes; and thus that the immigration and educationalpolicies behind them are anything other than substitutes.

13In a signalling model of education, as well, increasedimmigration in a particular occupation can lead to a de-cline in wages in that occupation and thus to a reduction

in the incentives to acquire the signal. If, however, thereis only a limited range of abilities that can be used in aparticular occupation, and there are no close substitutesin production for that occupation then there could be anargument in favour of using immigration policy if thereis an undersupply of the abilities associated with specificoccupations. On the other side, admitting immigrants withskill credentials that cannot be verified could lead to con-fusion on the part of employers about the value of thecredentials signal for all workers. This would imply areduction in efficiency in the labour market.

14In particular, an increase by 0.1 in the proportion ofa given (three year) entry cohort of immigrants who list aparticular occupation as their IO corresponds to only a0.05 increase in the proportion of that cohort observed inthat occupation at the next Census. Further, since the levelof occupational aggregation in Green (1995a) is substan-tial (only ten occupations are used), the amount of “leak-age” at the finer occupational groupings that have beenused for the selection system in the past must be larger.

15In a discussion with John Ries, he indicated that inmore recent work they find that there may be an entre-preneurial impact.

16Calculations from Labour Force Survey HistoricalStatistics Compact Disc.

REFERENCES

Abbott, M.G. and C.M. Beach (1993), “Immigrant Earn-ings Differentials and Birth-Year Effects for Men inCanada: Post War - 1972,” Canadian Journal of Eco-nomics 26(3):505-24.

Abella, I. and H. Troper (1983), None is too Many:Canada and the Jews of Europe, 1933-1948 (Toronto:Lester & Orpen, Ltd).

Akbari, A.H. and D.J. Devoretz (1992), “The Substitut-ability of Foreign-Born Labour in Canadian Produc-tion Circa 1980,” Canadian Journal of Economics25(3):604-14.

Avery, D. (1979), “Dangerous Foreigners”: EuropeanImmigrant Workers and Labour Radicalism in Canada,1896-1932 (Toronto: McClelland & Stewart).

Baker, M. and D. Benjamin (1994), “The Performance ofImmigrants in the Canadian Labor Market,” Journalof Labor Economics 12(3):369-405.

Bar-Or, Y., J. Burbidge, L. Magee and L. Robb (1995),



“The Wage Premium to a University Education inCanada, 1971-1991,” Journal of Labor Economics13(4):762-94.

Barro, R. and X. Sala-i-Martin (1992), “Regional Growthand Migration: A Japanese-United States Comparison,”Journal of the International and Japanese Economies(December):312-46.

Beaudry, P. and D.A. Green (1997), “Cohort Patterns inCanadian Earnings: Assessing the Role of Skill Premiain Inequality Trends,” Discussion Paper no. 97-03(Vancouver: Department of Economics, University ofBritish Columbia).

Blackorby, C. and D. Donaldson (1984), “Social Criteriafor Evaluating Population Change,” Journal of PublicEconomics 25:13-33.

Borjas, G.J. (1995), “The Economic Benefits From Im-migration,” The Journal of Economic Perspectives9(2):3-22.

Borowski, A. and A. Nash (1994), “Business Immigra-tion,” in Immigration and Refugee Policy: Australiaand Canada Compared, ed. H. Adelman, A. Borowski,M. Burstein and L. Forster (Toronto: University ofToronto Press).

British Columbia (1996), Work Futures (Victoria: Minis-try of Education, Skills and Training).

Canada (1961), Report of the Special Committee of theSenate on Manpower and Employment (Ottawa: Sup-ply and Services Canada).

_____ (1995), “Immigration Regulations, 1978: Amend-ment,” Canada Gazette, 18 November, pp. 3951-90.

Canada. Citizenship and Immigration Canada. (1994a),Annual Report to Parliament. Immigration Plan, 1994(Ottawa: Supply and Services Canada).

_____ (1994b), Into the 21st Century: A Strategy for Im-migration and Citizenship (Ottawa: Supply and Serv-ices Canada).

_____ (1996), A Broader Vision: Immigration, 1996 An-nual Report to Parliament (Ottawa: Supply and Serv-ices Canada).

Canada. Employment and Immigration Canada (1985a),Report to Parliament on the Review of Future Direc-tions for Immigration Levels (Ottawa: Supply andServices Canada).

_____ (1985b), Annual Report to Parliament on FutureImmigration Levels (Ottawa: Supply and ServicesCanada).

_____ (1992), Managing Immigration: A Framework forthe 1990s (Ottawa: Supply and Services Canada).

Canada. Health and Welfare Canada (1994), ChartingCanada’s Future: A Report of the Demographic Re-view (Ottawa: Supply and Services Canada).

Canada. Human Resources Development Canada (1994),News Release: Youth Internship Program (Ottawa:HRDC).

Card, D. and T. Lemieux (1997), “Adapting to Circum-stances: The Evolution of Work, School and LivingArrangements Among North Amercian Youth,” (Mon-treal: Université de Montréal, unpublished paper).

Coulson, R.G. and D.J. DeVoretz (1993), “Human Capi-tal Content of Canadian Immigrants: 1967-1987,”Canadian Public Policy/Analyse de Poli t iques19(4):357-66.

Danysk, C. (1995), Hired Hands: Labour and the Devel-opment of Prairie Agriculture, 1880-1930 (Toronto:McClelland & Stewart).

Denton, F., C. Feaver and B. Spencer (1997), “Immigra-tion, Labour Force and the Age Structure of the Popu-lation,” paper presented at the Canadian EmploymentResearch Forum Conference on Immigration, Employ-ment and the Economy, Richmond, BC.

Dougherty, C.R. (1994), “The Secondary Migration ofImmigrants to Canada” (Ottawa: Citizenship and Im-migration Canada, unpublished paper).

Economic Council of Canada (1991), New Faces in theCrowd: Economic and Social Impacts of Immigration(Ottawa: Supply and Services Canada).

Friedberg, R.M. and J. Hunt (1995), “The Impact of Im-migrants on Host CountryWages, Employment andGrowth,” The Journal of Economic Perspectives9(2):23-44.

Globerman, S. (1995), “Immigration and Trade,” in Di-minishing Returns:The Economics of Canada’s RecentImmigration Policy, ed. D.J. DeVoretz (Toronto: C.D.Howe Institute).

Green, A.G. (1976), Immigration and the Postwar Cana-dian Economy (Toronto: Macmillan).

_____ (1994), “International Migration and the Evolu-tion of Prairie Labour Markets in Canada, 1900-1930,”in Migration and the International Labor Market,1850-1939, ed. T.J. Hatton and J.G. Williamson (Lon-don: Routledge).

_____ (1995), “A Comparison of Canadian and US Im-migration Policy in the Twentieth Century,” in Dimin-ishing Returns: The Economics of Canada’s RecentImmigration Policy, ed. D.J. DeVoretz (Toronto: C.D.Howe Institute).



_____ (1996), “The Political Economy of Canadian Im-migration Policy, 1900-1930,” (Kingston: Queen’sUniversity, unpublished paper).

Green, A.G. and A.J. Green (1997), “The Implementa-tion of Immigration Policy in Canada,” (Kingston:Queen’s University, unpublished paper).

Green, A.G. and D. A. Green (1993), “Balanced Growthand the Geographical Distribution of European Immi-grant Arrivals to Canada, 1900-1912,” Explorations inEconomic History 30 (January):31-59.

_____ (1995), “Canadian Immigration Policy: The Effec-tiveness of the Point System and Other Instruments,”Canadian Journal of Economics 28(4b):1006-41.

Green, A.G. and G.R. Sparks (1999), “Population Growthand the Dynamics of Canadian Development: AMultivariate Time Series Approach,” Explorations inEconomic History 36:56-71.

Green, D.A. (1995a), “Intended and Actual Occupationsof Immigrants,”in Diminishing Returns:The Econom-ics of Canada’s Recent Immigration Policy, ed. D.J.DeVoretz (Toronto: C.D. Howe Institute).

______ (1995b), “The Early Adjustment of Immigrants:Credentials Recognition and Earnings Growth,” Dis-cussion Paper No. 95-12 (Vancouver: Department ofEconomics, University of British Columbia).

_____ (1999), “Immigrant Occupational Attainment: As-similation and Mobility Over Time,” Journal of LaborEconomics 17(1):49-79.

Hawkins, F. (1972), Canada and Immigration: PublicPolicy and Public Concern (Montreal and Kingston:McGill-Queen’s Press).

Head, K. and J. Ries (1998), “Immigration and TradeCreation: Econometric Evidence from Canada,” TheCanadian Journal of Economics 31(1):47-62.

Helliwell, J.F. (1998), “National Borders, Trade and Mi-gration,” NBER Working Paper No. 6027 (Cambridge,MA: National Bureau of Economic Research).

Iacovetta, F. (1992), Such Hardworking People: ItalianImmigrants in Postwar Toronto (Montreal and King-ston: McGill-Queen’s University Press).

Kunin, R. and C.L. Jones (1995), “Business Immigrationto Canada,” in Diminishing Returns: The Economicsof Canada’s Recent Immigration Policy, ed. D.J.DeVoretz (Toronto: C.D. Howe Institute).

Leacy, F.H., ed. (1983), Historical Statistics of Canada,2d ed. (Ottawa: Statistics Canada).

Lin, Z. (1997), “Foreign-Born vs. Native-Born Canadi-ans: A Comparison of their Inter-Provincial Mobility”(Statistics Canada: Business and Labour MarketAnalysis Division).

Pope D. and G. Withers (1993), “Do Migrants Rob Jobs?Lessons of Australian History, 1861-1991,” The Jour-nal of Economic History 53(4):719-37.

Piva, M.J. (1979), The Condition of the Working Class inToronto, 1900-1921 (Ottawa: University of Ottawa Press).

Ramirez, B. and M. Del Balso (1980), The Italians ofMontreal: From Sojourning to Settlement, 1900-1921(Montreal: Associazione di Cultura Popolare Italo-Quebecchese).

Richmond, A.H. (1967), Post-War Immigrants in Canada(Toronto: University of Toronto Press).

Simon, J.L. (1989), The Economic Consequences of Im-migration (Oxford: Basil Blackwell Ltd.).

Timlin, M.F. (1960), “Canada’s Immigration Policy, 1896-1910,” The Canadian Journal of Economics and Po-litical Science 26(4):517-32.

Trefler, D.(1997), “Immigrants and Natives in GeneralEquilibrium Trade Models,” NBER Working Paper6209 (Cambridge, MA: National Bureau of EconomicResearch).

Wright, R.E. and P.S. Maxim (1993), “Immigration Policyand Immigrant Quality: Empirical Evidence fromCanada,” Journal of Population Economics 6(4):337-52.

The Economic Goals of Canada’s Immigration Policy: Past...

Documents

Transcript of The Economic Goals of Canada’s Immigration Policy: Past...