LABOUR SHORTAGES AND THE NEED FOR IMMIGRANTS: OECD
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Part II
LABOUR SHORTAGES AND THE NEED FOR IMMIGRANTS:
A REVIEW OF RECENT STUDIES1
Introduction ing participation rates, postponing retirement ages,
stimulating the labour market participation of
Economic, demographic and political develop- women and immigrants are among those instru-
ments, combined with the mounting concern about ments, as is facilitating selective employment-
the future of labour supply have renewed the atten- related immigration policies. Whilst each of these
tion of policy-makers on labour migration issues. A instruments may contribute, no single instrument in
good economic climate and years of economic itself can promise the complete solution. However,
growth in some OECD countries have led to increas- faced with the urgent need to meet a growing
ing employment, higher participation rates and demand for workers with specific skills in certain
lower unemployment rates. Employers started to sectors, some countries have already started to facil-
experience difficulties in filling vacant posts, con- itate labour market access for skilled immigrant
cerns about the availability of labour on domestic workers. Some countries have considered recruit-
markets arose and the first calls for immigrant labour ment of immigrants on the basis of their human
echoed. capital, i.e., through selective immigration, to com-
The increasing diversity of migrant’s nationali- pensate for an ageing workforce. Hence, the policy
ties and of the migration channels used, as well as re le vance of considerin g the comple te set of
the growing proportion of movements of temporary options as well as the advantages and limits of
and skilled workers in total migration flows reflect selective labour immigration.
the influence of increasing internationalisation on The aim of this chapter is first to show how a
international migration. Migration has not, however, number of OECD member countries estimate the
accelerated at a pace comparable to trade flows and current and forecast labour shortages (A). The role of
capital movements (Tapinos and Delaunay, 2000). migration in alleviating future labour market needs
Demographic developments in OECD coun- will be then presented (B). Section C will review sec-
tries imply an ageing workforce and ultimately result tor specific and micro-level studies making the link
in a declining population of working age. The possi- between labour shortages and the need for selec-
bility that these developments also result in labour tive employment-related immigration policies.
shortages at the macro-level has added an extra
dimension to discussions on the need for immi- A. MEASURING CURRENT LABOUR
grants. In the medium term, as early as 2015 for SHORTAGES
some countries, the increasing number of retiring
baby-boomers will in some occupations lead to a Labour shortages are not easy to measure. At
replacement labour demand that may be hard to fill the macroeconomic level, reference is made to fall-
from domestic labour supplies. Some countries with ing unemployment rates and increasing employ-
low labour force participation rates may be able to ment, sometimes coupled with accelerating real
use their large labour reserves. earnings to indicate growing labour market tight-
Policy makers have various instruments avail- ness. However, labour market tightness is better
able to react to potential labour shortages. Increas- measured by comparing actual unemployment rates 103
© OECD 2003Trends in International Migration
with structural rates of unemployment; frictional and 2. Causes of labour shortages
structural labour market rigidities can be a cause of
persisting unemployment combined with wage infla- Labour shortages have a variety of causes. They
tion. Because structural rates of unemployment are may occur because of a lack of geographical mobility
not readily available, unemployment rates are on the part of the workers. They may be the result of
sometimes given in relation to the vacancy rates for a mismatch between qualifications. For example,
a specific occupation (defined as ratio of the num- there is evidence of a mismatch between the low
ber of vacancies and the number of employed in qualifications of the numerous unemployed and
that occupation). Zimmerman et al. (2002) have in a cti ve wo rk e rs a nd t h e h ig h qu a lifi cat io n s
calculated Beveridge curves for 1980-95 for 40 occu- needed in Germany’s technology intensive industry
pational groups in Germany.2 They have found indi- (Winkelman, 2001; Munz and Ochel, 2001; Süssmuth,
cations of shortages in the early 1990s for engineers, 2001). Zimmermann et al. (2001) find evidence of
stone masons and technicians, persisting until 1995 mismatch-unemployment in shifting Beveridge
for health-related occupations. In France, the occu- curves for welders, technicians, electricians, occupa-
pational job-seekers ratio, defined as the ratio of tions in the food industry and construction workers,
the number of job-seekers in a specific occupation carpenters and painters. In other cases, the occupa-
to the total of job-seekers and employed workers in tion’s image is not good enough, at the going wage
that occupation, provides a disaggregated illustra- rate, to attract a sufficient number of workers, even if
tion of the tightness in the labour market. It indi- supply is abundant. Klaver and Visser (1999) find
cates increasing tightness in the labour market in this effect for different sectors in the Dutch econ-
construction and mechanical industries (DARES, omy. Among the other important factors explaining
2001a and 2001b).3 labour shortages are demographic factors, rapid
technological progress and the resulting change in
employment structure towards highly skilled labour
1. Trends in national statistics on the number cyclical variations in demand, production and the
of registered vacancies by sector derived labour demand, rigid wage structures and
of employment and occupational category collective agreements (Munz and Ochel, 2001;
Australian Productivity Commission, 2000).
Increases in the number of vacancies in and
outside the high-tech sectors 4 in non-European The Netherlands, for example, argue that labour
OECD countries give similar indications of strongly shortages would develop less if the number of sepa-
increasing numbers of unfilled job vacancies, con- rations were reduced5 (Tweede Kamer, 2001b). The
centrated in a small number of not necessarily Belgian Federal Planning Bureau (FPB) finds in the
highly-skilled occupations. For example, the Austra- absence of a positive wage drift and inflation the
lian vacancy bank Job Search listed as the top four indication that general labour shortages are neither
most wanted occupational groups among the total of an issue now nor in the medium term. However,
40 500 job vacancies in mid December 2001: Labour- demographic developments may change the situation
ers, Factory and Machine Workers (9 000); Food, for the longer term (FBP, 2001). Austria considers
Hospitality and Tourism (5 600); Sales Assistants labour shortages as a consequence of mismatches in
and Storepersons (4 200); and Gardening, Farming the labour market, but takes the demographic
and Fishing (3 700) (DEWR, 2002). developments into account when predicting the
medium term consequences (Biffl and Walterskirchen,
Another possible measure of labour market 2001). The various explanations of the existence of
shortages is found in employers’ reports where current labour shortages and the potential effects of
th e y e xpr ess th e ir con ce rn a bou t th e co n se - demographic trends result in a number of different
quences of developing shortages of qualified definitions of labour shortages.
workers in certain occupations. The surveys con- To sum up, the different causes of labour short-
firm the picture that vacancy data give: shortages ages identified from the studies discussed in this
appear to be concentrated in specific sectors but chapter can be summarised as follows:
these sectors are not the same across countries.
The results also show that employers experience • Continuing technological change may lead to
shortages mainly at higher skill levels, but not a structural shortage of workers in possession
104 exclusively (see Annex II.1). of the needed skills: workers neither had the
© OECD 2003Labour Shortages and the Need for Immigrants: A Review of Recent Studies
time nor the opportunity to invest in these The need for immigrant labour depends in
skills. large part on a country’s initial situation. Countries
• Slow adjustments in the labour market may with low unemployment rates and high activity
cause shortages. It takes time for employers rates, like Norway, might exhaust earlier opportuni-
to recognise labour shortages and to react to ties to retrain unemployed workers and inactive
them, for example by offering higher wages. It people than countries with a more abundant stock
also takes time for workers to recognise bet- of people not in work. Two factors play a determin-
ter opportunities elsewhere and to react to ing role: the size of the potential labour force and
them. Employers may be reluctant to raise the obstacles to labour-market integration that cer-
wages or are tied to collective agreements or tain groups face.
inflexible remuneration structures. The resident potential labour supply consists of
• Mismatch: wrong education investment deci- unemployed people, people available but not par-
sions resulting in too few engineers, scientists ticipating in the labour market and people not avail-
and doctors, for example. able and not participating. There also is potential
• Insufficient regional labour mobility. labour supply among people who are employed
involuntarily in a part-time job. The group of unem-
• Institutional or demographic causes: a high ployed people can be further divided into recently
number of people in retirement or invalidity unemployed people and people who have been
pensions, low female participation rates. unemployed for a longer period of time. The former
There is no universally applied definition of could be referred to as frictional unemployment,
labour shortages. In part this reflects the fact that which is caused by a number of people in the pro-
they are not easy to measure. Macroeconomic data cess of changing jobs.
do not reveal that tightness can be severe for some
occupations whilst absent in others. Employer The second group of people, the non-
reports and surveys confirm that shortages occur in participants, is a much more heterogeneous group
specific occupations and at different skill levels. than that of the unemployed. Other than the unem-
However, employers report on recruitment difficul- ployed, not all non-participants are available for
ties but not on labour shortages per se. The studies work – they may be facing serious obstacles. A con-
discussed in this chapter define labour shortages in siderable part of this group consists of people
different ways to explain their possible different whose health does not allow them to work. Another
causes and forecast their potential development. part is formed by people who are otherwise not
available for work either because care-giving tasks
occupy their time or simply because they do not
3. Labour reserves exist but it would be difficult
wish to work. The unavailability for work may be
to mobilise them rapidly
temporary, as for students and others temporarily or
The arguments in favour of attracting immigrant partly inactive because of education or professional
workers to solve labour shortages rest partly on the training.6 But there are also a number of people who
assumption that the resident labour supply is insuf- are in principle available for work although they are
ficient, or of inferior quality, to meet labour demand. not actively looking for work.7
But if the number of resident unemployed and inac-
The size of the unused labour supply varies
tive people can largely fill labour demand in quanti-
enormously, ranging in 2000 from 18 to 22% in
tative terms, then why not proceed that way? The
Switzerland and Sweden to 40% in Italy (OECD,
answer to this adds considerably to the sensitivity of
2001e). Some research argues that for countries like
the issue of labour migration. Some governments
the Netherlands (Tweede Kamer, 2001a and 2001b)
hold the view that the unused resident labour sup-
and Belgium (Feld, 2002) latent labour supplies are
ply is sufficient to solve the present and future
such that general labour shortages are not likely to
labour shortages – for example, Finland and the
develop.
Netherlands – and do not envisage any changes to
the existing restrictions on labour immigration rules. The OECD has recently started to investigate
At the same time, both these countries, like many the potential labour supply.8 Based on the observa-
others, facilitate the immigration of highly skilled tion that on average for the OECD slightly more than
workers for occupations where the national labour 55% of the working age population in 1999 was in
market cannot supply sufficient labour. full-time employment, the OECD report aims to 105
© OECD 2003Trends in International Migration
indicate the potential for full-time employment rates on the one hand are very low – well below the
among people in part-time employment, unemploy- replacement level of 2.1 children per woman – and
ment and inactivity. The remainder, 45%, is either falling. On the other hand, people’s life expectancy
(partly) unemployed or (partly) inactive and, in the- at birth continues increasing, which goes together
ory, represents the latent labour supply. Unem- with improving prospects of a healthy old age. The
ployed people accounted for 5% and the biggest combination of low fertility and increasing longevity
chunk, more than 27% of the entire population of means that population growth rates are slowing
working age, was not in the labour force for various down and that there is an increasing proportion of
reasons. elderly in the total population.
Labour Force Survey questions on the reasons The prospect of rising shares of the elderly in
for not looking for work are subject to a certain the population and falling shares of the population
amount of interpretation. For example, respondents of employment age to support pension and health
referring to family responsibilities may mean that systems has been an issue of previous OECD work
they are unable to free themselves of these respon- (OECD, 1998a and 1998b). Population ageing will
sibilities because of non-availability of child-care or have far-reaching effects, some of which are of par-
because they do not wish to work under current con- ticular importance for the discussion of the role of
ditions. The availability of child-care could be the migration policies:
object of a policy initiative whereas changes in the • Demographic developments will change
individual’s personal situation or in labour market or labour supply. The labour force changes in
policy inducements to participate could change his/ composition as the share of older workers
her preference to work. In this regard, the answers to grows, and it may dwindle if people continue
the questions on the desire to work would help dis- retiring early. The number of new entrants
tinguish “real” impediments from what could be into the labour market may be insufficient to
termed “competing activities” (OECD, 2002d). How- replace retiring workers.
ever, respondents who indicate no desire to work
may well change opinion, for example students in • The ageing of the workforce may adversely
their final year. affect macroeconomic performance if future
labour supply is inadequate to meet the
A considerable portion of the working age pop- qualitative and quantitative need for workers
ulation is not actively engaged in any form of to sustain economic expansion or preserve
employment or study. This is referred to as the resi- present welfare levels.
dent latent labour force. To the extent that govern-
ments succeed in mobilising this unused potential, • The growing number of people relying on
current labour shortages can be alleviated, reducing retirement benefits will shift the balance
the need for immigrant labour. The analysis shows between working people (contributors) and
that the most important potential is among middle- inactive people (beneficiaries) to an extent
aged inactive women, most of whom have less than that may put the finance of social security sys-
intermediary qualifications. Another important tar- tems at risk.
get group is youth, a state of things caused by a mis- The role of migration in addressing these chal-
match between education out-turn and labour lenges, in particular the third issue of financing
demand. Further work is needed to obtain timely social security, was the subject of a recent UN report
quality labour force data for the OECD countries. (UN, 2000). The report concluded that keeping old
More analysis is also needed to establish the type age dependency ratios at current levels through
of education and recent experience of the latent migration alone seems out of reach because of the
supply. extraordinarily large numbers of migrants that would
be required. The study presented different scenar-
B. THE ROLE OF MIGRATION IN ADDRESSING ios with regards to international migration streams
FUTURE LABOUR MARKET NEEDS needed to achieve specific population objectives:
to maintain the size of the population and of the
1. Demographic challenges working population or to maintain the old age
dependency ratio at a constant level. The study
Two phenomena mark current demographic estimated that to keep the old age dependency
106 trends in most industrialised countries. Fertility ratio constant at the highest level it would reach in
© OECD 2003Labour Shortages and the Need for Immigrants: A Review of Recent Studies
the absence of migration after 1995, “for the EU Two important observations moderate the
alone a total number of 674 million migrants over extreme hypothesis of the UN report. First, the
the 2000 to 2050 period, or 13 million per year” choice of time horizon is essential to the evaluation
would be needed. In comparison, the 1998 revision of the demographic effects of migration and, more
Population Prospects (medium variant) on the basis relevant to the present context, of the effects of
of current flows estimated the total over the entire demographic developments on the labour market.
period at 13.5 million or a yearly average of 270 thou- The second observation is that the UN uses its pro-
sand (UN, 2001). jections of the old age dependency ratios on the
assumption that labour force participation and
Despite its extraordinarily high estimates the
retirement ages do not change. The development of
UN report has attracted a lot of media attention and
the size of the labour force is under those assump-
certainly has helped to put the discussion of labour
tions proportional to the size of the working age
migration issues back high on the political agenda.
population as shown in Chart II.1. Chart II.1A shows
Increased immigration has the advantage of having
that Australia, New Zealand, Ireland, Canada, and
an immediate effect on the age and composition of
the United States are exempted from the trend of
the population because of the younger age structure
declining working age populations. In many other
of net migration. In addition, fertility rates of immi-
countries, the working age population will decline
grant women are often relatively high, which can
rapidly and will fall well below its present levels
boost fertility and hence long-term population
by 2030. Chart II.1B shows a number of OECD coun-
growth. However, the role of migration policies in
tries with decreasing working age populations. The
addressing these challenges can only be comple-
countries are grouped by the size of their popula-
mentary to other policies and is subject to a number
tions. The working-age population will decrease,
of practical and political constraints (OECD, 1991).
according to UN projections, slowest in France (–8%
Tapinos, for example, noted with reference to by 2050) and fastest in Italy (–42% by 2050). In the
the UN report that further analysis of the practical Nethe rlands, the working-age populatio n will
and political issues involved in the implementation decrease by 14% by 2050, in Austria by 36%. The
of an effective demographic and labour market ori- decline of the working age population appears to be
ented policy is needed. This would enable OECD relatively moderate in Belgium, Denmark and the
member countries to have a clearer picture of the Netherlands, the United Kingdom, Norway and
extent to which international migration can contrib- France, but rapid in most other countries.
ute to positive outcomes (Tapinos, 2000). He quoted In this context, Tapinos notes that the old age
different reasons why immigration cannot be a sole dependency ratio is a demographic ratio that com-
solution to population ageing: pares age groups and should more realistically be
• Most OECD countries share the same demo- replaced by the ratio of retired persons to employed
graphic pattern and the bulk of immigration persons (Tapinos, 2000). Depending mainly on
would have to come from outside the OECD female participation, average retirement ages and
area. on employment rates, there may be a significant dif-
ference between the two ratios. The following dis-
• Migration cannot be treated as a control vari- cussion shows how some countries take account of
able because there are too many factors that expected changes in labour force participation rates
limit the ability to control and select migra- and retirement ages in the forecasting of labour
tion flows. In addition, migration policies force developments.
have limited effects on return migration.
• Because the means available to monitor the Migration and sustaining the labour force9
trend of net migration differ across countries,
Ba se d o n t h e U N P op ul at io n P ro j ect io n s
it is unclear whether migration policy targets
of 1999, the Council of Europe has analysed the
are actually attained.
effects of demographic changes in the working age
• Finally, immigrants have a better chance of population on employment and unemployment
integrating in a society that experiences a nat- (Council of Europe, 2000). The study addresses the
ural increase, which reduces the value of question of how and to what extent labour migration
immigration as a substitute for declining fer- may be affected by labour shortages, thereby taking
tility rates. into account the possibility that activity rates change 107
© OECD 2003Trends in International Migration
Chart II.1. Projections of the working age population (15-64) in selected OECD countries
A. Countries with working age populations increasing by 2050
2000 = 100
New Zealand Ireland Australia Canada United States
130 130
125 125
120 120
115 115
110 110
105 105
100 100
2000 2030 2050
B. Countries with working age populations decreasing by 2050
Thousands
2000 2030 2050
100 000 100 000
80 000 80 000
60 000 60 000
40 000 40 000
20 000 20 000
0 0
Japan Germany Italy United Kingdom France Spain Poland
12 000 12 000
10 000 10 000
8 000 8 000
6 000 6 000
4 000 4 000
2 000 2 000
0 0
Netherlands Portugal Belgium Sweden Austria Switzerland Denmark Finland Norway
108 Source: United Nations, 2001.
© OECD 2003Labour Shortages and the Need for Immigrants: A Review of Recent Studies
over time and across countries. It does not, however, Kingdom, but not in the other countries. This could
take account of the effects of possibly changing imply a need for more migration, except in the cases
actual retirement ages. Up to 2010, the size of the of Ireland and Portugal, where the demographic
labour force is projected to increase in all countries. effects are positive.
Projections are given up to 2020 for France, Poland, Feld concludes, however, that the evaluation of
the Netherlands and Austria; a declining labour the role of migration in attenuating the expected
force can be seen in Austria, France and Poland. declines of the labour force has a number of meth-
Chagny et al. (2001) forecast the development of odological and political problems. Forecasts of
the European labour force, taking into account the labour force developments assume that migratory
effects of migration and changing participation rates. flows in respect of age composition and participa-
Allowing for an annual immigration that rises from tion rates are identical to the host country popula-
7 00 0 00 pers ons in the 19 95 -2 00 0 period t o tion and neglect the fact that net migratory flows
900 000 persons by 2020, the authors expect the involve inflows and outflows that may vary hugely in
European labour force to decline rapidly after 2015. composition and magnitude (see also Box II.1).
Taking the analysis of the interactions between Another issue is that situations of neighbouring
demographic trends and the labour force one step countries may vary considerably; labour shortages in
further, Feld (2001) looks at 15 EU countries for the some countries may occur 10 or even 20 years
period of one generation (25 years) ahead. The before they emerge in other countries.
author specifically addresses the question whether For Scandinavian countries, the Nordic Council
population ageing will lead to labour shortages and, of Labour Ministers recommends a number of
if so, whether large-scale immigration will be neces- actions to deal with the expected decline of the
sary. Bringing together information from different labour force. First and foremost, they intend to
countries using a common methodology, Feld calcu- mobilise the latent labour supply among four target
lates when the labour force in the EU countries will groups: the ageing population; inactive and unem-
fall back to its 2000-level. He distinguishes between ployed youth; inactive adults and inactive and
demographic effects (fertility, life expectancy and unemployed foreign-born residents. They do not,
migration) and behavioural effects (participation however, give quantitative estimates of the poten-
rates, retirement age, size of the school-going popu- tial impact of any of these measures but compare
lation). According to Feld’s estimate, behavioural best practices among Nordic countries. One of its
effects can compensate for demographic effects in conclusions confirms the findings for the EU quoted
Denmark, France, the Netherlands and the United above (increased migration might contribute to
Box II.1. Gains and losses in high skilled labour through migration
Migration involves inflows and outflows that may vary hugely in composition and magnitude. Some countries
have recently drawn up the balance of skill gains involved in immigration and skill losses involved in emigration.
As a result of the international movement of skilled workers to and from Australia over the five-year
period 1995-1996 to 1999-2000, Australia registered a net “brain gain”. There was a net loss of skilled residents
but this loss was more than offset for almost every occupation by gains from permanent and temporary
immigration. The lack of quality data leaves open the possibility that Australia is losing high quality residents
and replacing them with lower quality settlers and visitors, albeit in greater numbers (Birrell et al., 2001).
Canada suffered a net loss of skilled workers to the United States in several economically important
occupations during the 1990s, although the numbers involved have remained small in an historical sense and
small relative to the supply of workers in these occupations. (See also Bordt in OECD, 2002b.) However, whilst
losses of highly skilled workers to the United States accelerated during the 1990s, so too did the influx of highly
skilled workers into Canada from the rest of the world. The balance is neutral: the number of master’s and
doctoral graduates entering Canada from the rest of the world is equal to the number of university graduates at
all levels leaving Canada for the United States (Zhao et al., 2002).
109
© OECD 2003Trends in International Migration
shifting the trend of an ageing workforce). Restricting the demographic change is limited compared
the immigration of people with low qualifications to to other EU countries and because it has a
prevent integration difficulties is among the policy large second pillar of privately funded occu-
options; the Nordic Council of Labour Ministers cau- pational pension schemes (Van Ewijk et al.,
tions that in this case integration policies need fur- 2000). However, the study makes some assump-
ther attention (Nordic Council, 2000). tions about participation rate, life expectancy,
productivity growth and world interest rates
At the level of individual European countries, (which might not hold), and fiscal policies
studies of demographic impacts on the labour mar- alone may not be enough to achieve budget
ket often confirm the expected trends summarised neutrality. The CPB assumes that the bulk of
in the preceding paragraphs. Some of them illus- immigrants will come from Third World coun-
trate the role migration could play in combination tries and argues that because their current
with other policies. participation rates are much lower than for
natives, they are much more likely to be ben-
• For France, Blanchet (2001) finds contradictory efit dependent and add to the stock of inac-
evidence in a literature review of demographic tive persons living off public money.
impacts on trends in wages, employment and
unemployment. Domingues Dos Santos (2000) The Australian Treasury directly links the
looks into the budgetary effects. She argues growth of working-age population to economic
that to maintain the financial sustainability of growth, thus implying positive effects of net immi-
the social security system (i.e. pension system) gration (Australian Treasury, 2000). Withers (1999)
with the given unemployment rate, either the demonstrates that a proper estimation of migration
retirement age should be raised by 11 years, rates should include dependency ratios and related
benefit levels reduced by 14% or contribution budget costs. This can result in an argument in
rates increased by 4.5%. Whilst both papers favour of much higher immigration; it would have
discuss the demographic impacts on the the effect of halving the public cost of an ageing
labour market, they do not refer to migration population. However, McDonald and Kippen (2001)
as a policy alternative. argue that the budgetary effects of higher migration
• The German Institute for Employment are not clear because the immigrant population also
Rese ar ch (IAB) e stimat es th e po te nt ial ages. In their estimates, net immigration of about
domestic labour supply from 1996 to 2048 80 000 persons a year makes a reasonable contri-
under different assumptions about female bution to the reduction of the ageing of the popu-
participation rates and higher immigration lation. Higher numbers are less effective, mainly
(Fuchs and Thon, 1999; IAB, 1999). IAB argues because the immigrant population also ages. By
that German labour supply will decline, what- comparison, Australia’s net immigration has ranged
ever policy is chosen. However, high immigra- between 80 000 and 107 000 between 1995 and 1999
tion does postpone the year when labour (OECD, 2001d).
supply will start to decline, especially in com-
For the United States, the Centre for Retire-
bination with an increase in labour participa-
ment Research finds that men and women between
tio n rates. For example, in the scenario
25 and 54 years old account for 71.7% of the labour
whereby female participation rates catch up
force in 1998 but expects that their share will fall to
with male participation rates and net immi-
68.1% as early as 2006 (Hammermesh, 2001). As the
gration is zero, the labour force would start to
oldest baby boomers will begin to retire, the impli-
decline by 2000. If in a similar situation,
cation s for t he workfo rce cou ld be en ormo us
500 000 immigrants per year as of 2000 are
(Gottschalk, 2001). The current tight labour market
admitte d, th e o nset o f the labo ur fo rce
situation could be exacerbated, hindering prospects
decline could be postponed by 20 years
of economic growth and putting a greater burden on
to 2020.
those remaining in the workforce, perhaps forcing
• The Netherlands Bureau for economic policy them to work longer hours. After 2008, as more and
analysis (CPB) established on the contrary more baby boomers reach retirement age, the
that ageing will have fairly modest economic impact of their retirements will continue growing
110 consequences in the Netherlands, because (Dohm, 2000).
© OECD 2003Labour Shortages and the Need for Immigrants: A Review of Recent Studies
2. Special studies on the economic need adjust to immigrant labou r. If migration
for migration started with 100 000 yearly, implying a cut in
migration compared to the actual situation,11
The developments raised in this chapter, com- increasing to 150 000 a year by 2020 and then
bined with the mounting concern about the future of further to 300 000 a year the active labour
labour supply have triggered a number of special force would decline least and economic
studies of the economic effects of international growth would be maximised.
migration and the need for foreign labour. The moti-
vation for each of the studies is to provide a basis The activities of the Süssmuth Commission
upon which policy-makers can decide on the need have led to an entirely new and coherent immigra-
to review migration practices and policies and on tion law that should come into force on the 1st of
the manner in which eventually such revisions could January 2003. This law establishes clearly the rights
be made. It appears that employer reports of a of foreigners to remain and work in Germany and
shortfall of qualified labour supply and studies on simplifies the existing law. As to labour migration,
the demographic impacts on the labour force and the law distinguishes high-skilled from low-skilled
labour market have not been sufficient. workers. Only the highly-skilled are eligible for a
permanent permit. However, where shortages exist,
Germany the decision to resort to temporary immigrants can
be made at a regional level. The law refers to a
Th e G e rma n co mmissio n o n imm igra tio n points-system, yet to be developed, as an addi-
(Süssmuth Commission) has undertaken a study of tional instrument to attract immigrants. Such a sys-
the role of migration in the economy, now and in tem would be subject to separate approval by the
future (Süssmuth, 2001). Among its main findings German Länder.
are:
• Large pools of unused labour reserves con- The United Kingdom
sisting of unemployed and inactive people
co-exist with increasing labour demand. Lack The Home Office, in charge of immigration and
of qualifications is the main reason for the integration, has published a study aiming to contrib-
unemployment of around 4 million people. ute to: “a sound understanding of the impact of
However, the report does not give a clear existing policies affecting migration and migrants,
answer as to the extent to which the present and a framework for assessing the costs and benefits
labour demand can be met through activating of potential alternatives” (Glover et al., 2001). The
and upgrading the skills of the unemployed. study looks at migration “in the round”; proceeding
with a discussion of the Government’s higher level
• Immigration can contribute to maintaining objectives and examining the social and economic
present economic welfare levels conditional outcomes that current policy delivers. The study
on the good integration of migrants. Better embraces the idea that migration enables industries
qualified and younger migrants have more to expand. Using the information technology (IT)
im me diat e a nd mo r e size a ble po sit iv e industry as an example: “so if migration of workers in
effects. particular sectors is restricted […] then it will not be
• If Germany maintains a net immigration at primarily the case that supply of, and wages of
200 000 yearly – the average level for the past native British IT workers will increase. The IT indus-
20 years – the decline of the German popula- try will simply shrink relative to what would have
tion will be limited to 12 million people. happened with a less restrictive policy.” The esti-
Without immigration, it will shrink by 23 million mate is that a 1% population growth through migra-
people by 2020 compared to 1999/2000.10 The tion could be associated with a 1.25 to 1.5% increase
share of older people in the population (rela- in GDP for European countries.
tive ageing) will increase whatever the migra- Like the German report, the Home Office study
tion scenario, because migrants age as well. is an ad-hoc report that focuses not only on the
• Immigration has significant effects on wage- need for foreign labour and migration policy but
flexibility and investment in education. analyses the past and present effects of migration as
Therefore immigration intake should start at well. Both studies result in a number of policy rec-
modest levels to give the market time to ommendations. But whereas the German study is 111
© OECD 2003Trends in International Migration
very detailed and extensive, and definitive in that it • Austria has become a net-immigration country
has led to a proposal for a new immigration law, the without developing the instruments for an
UK study serves as a basis for further work. It does immigration policy. The basics for a targeted
not address any of the issues related to the demo- immigration policy do not exist yet. Inconsis-
graphic impact on labour supply nor a humanitarian tent data about the foreign population do not
immigration and family reunion, which the German give sufficient insight into the resident labour
report covers, and hardly touches on the need for supply. Acknowledging the potential contri-
good integration policies. However, the study also bution that a qualifications-oriented migra-
fits into a broader government-commissioned tion policy can make to solve many economic
framework making use of existing infrastructure to and social problems may ease the introduc-
improve the documentation on current and future tion of such policies.
labour market bottlenecks, analysing why compa-
nies ask for workers with new skills and – ultimately
– assessing labour demand and skill needs (DFES, Norway
2001c; Haskel and Holt, 1999).
Against the background of tight labour markets
Austria in 1997 and 1998, the Norwegian Ministry of Local
Government and Regional Development, which has
A particularly sizeable study by the Austrian the responsibility for immigration and integration,
Institute of Economic Research (WIFO), co-funded commissioned a pan-departmental task force with a
by the European Social Fund (ESF), the Ministry of study of labour market needs and recruitment pos-
Economics and Labour (BMWA) and the Ministry of sibilities from abroad (Kommunal-og Regionalde-
the Interior (BMI) covers demographic challenges, partementet, 1999). The task force proceeded in two
the educational attainment of foreign-born residents, steps. First, it identified 11 strands to immigration
foreigners’ labour market integration, crowding-out policy; second, it made recommendations as to the
effects on the labour market, family reunion, social feasibility of each of these strands. Their final report
costs and macroeconomic aspects of immigration. gives very little attention to the evaluation of cur-
Among the main conclusions are (WIFO, 2001): rent labour shortages, other than stating that short-
ages can be observed but cannot be forecast.
• Net immigration can contribute to sustained
Depending on the world-economy, however, and
economic growth and to solving problems
given the near-full utilisation of Norwegian domestic
related to the sustainability of the social secu-
labour supply, present shortages are likely to per-
rity system, next to a better use of the resident
sist, although they may well shift from one sector to
labour potential. However, current migration
another. Demographic developments will inevitably
policies are not adequate to deliver these con-
lead to growing labour market tightness. The main
tributions because family reunion and humani-
text appears to take the need for foreign labour for
tarian immigration play a dominant role, while
granted, although the magnitude may vary according
labour immigration is marginal.
to the scenario.
• Increasing participation rates of the sizeable
foreign-born resident population will only The report stresses the fact that labour immi-
have limited success because their qualifica- gration inevitably involves the immigration of peo-
tion structure is too much concentrated on ple with differing needs for housing, public services,
skills that are in low demand. The integration etc. Some options or decisions may cause a conflict
of immigrants’ children could be accelerated with the leading principle in Norwegian politics that
by le ga lisin g accompa nyin g families as all citizens should have equal opportunities, rights
quickly as possible, thereby helping them in and obligations. What to do, for instance, with tem-
establishing an education career and helping porary residents? The task force recommends facili-
Austrians in building a higher skilled labour tating the entry of high-qualified specialists and
force. However, the Austrian education sys- seasonal workers. This needs relatively few adjust-
tem is as yet not set up to train the relatively ments to immigration law and may offer a flexible
small cohorts of the future labour force for framework in reaction to the ever-changing labour
tho se qualificatio ns that are in growin g demands. It will have many positive and few nega-
112 demand. tive side effects.
© OECD 2003Labour Shortages and the Need for Immigrants: A Review of Recent Studies
The Netherlands mends selective and temporary labour migra-
tion (WRR, 2001a).
The major government institutions in th e • The Social and Economic Council (SER), com-
Netherlands have produced a vast body of research missioned by the Minister of Social Affairs
covering microeconomic effects of immigration, the and Employment, has looked into possibili-
demographic challenges as discussed above, issues ties for improvement of the low intra-EU
of social and labour market integration and case labour migration and to analyse the potential
studies of immigration experiences in neighbouring benefits for the Dutch labour market, in par-
countries and in the United States. ticular in alleviating current labour shortages
(SER, 2001). The study identifies obstacles in
• The Netherlands Bureau for Economic Policy
legislation that should enable EU-citizens to
Analysis (CPB) for the first time since 1972 has
reside and work in an EU-country of their
estimated the labour market effects of migra-
choice. But more importantly, the SER has
tion for the Netherlands (Roodenburg and
found that the access to certain professions is
Van den Boom, 2000). They find that the inter-
limited for citizens from other EU-countries.
national evidence need not necessarily hold
Examples are jobs where certain certificates
in Holland among other things because old-
are required that may not be obtainable in
age pensions are largely capital funded and
other EU countries, the incapacity of employ-
more or fewer migrants will hardly effect its
ers to judge the value of foreign degrees and
affordability. However, they also argue that
certain government-jobs that are closed to
the social security system is relatively gener-
foreigners. The SER also makes recommenda-
ous compared with that of the United States
tions to better enable job-seekers to orien-
(where most evidence is based) and costs/
tate themselves in other EU job-markets.
benefit ratio may well be negative. They
observe that the high population density in
the Netherlands may cause much more seri- C. STUDIES ON LABOUR SHORTAGES
ous congestion effects than elsewhere. AT MICRO-LEVEL AND THE NEED
FOR IMMIGRANT WORKERS
• The Dutch Scientific Council for Government
Macroeconomic studies may establish the posi-
Policy (WRR), considering that the Netherlands
tive overall effects of migration, the consequences
has developed into an immigration country,
of ageing and the effect of raising participation rates
h as initiate d a n ew stu dy on th e Multi-
and lowering retirement ages on labour supply; they
Cultural Society (WRR, 2001a). In the context
cannot explain on what migrant-characteristics
of the study, the WRR has asked several
selective immigration policies could be formulated.
researchers to compile the information neces-
Nor can they explain how many immigrants could
sary to form a solid empirical foundation for
and should be recruited to meet labour market
policy advice. The issues covered mostly deal
n e e ds a n d fo r h o w l o n g th e s e n e e ds ca n be
with the integration of foreigners into society,
expected to last. These questions are important
ranging from the institutions of Islam in the
against the background of the general tendency,
Netherlands and the demographic profile of
particularly in European and Scandinavian coun-
f o r e i g n bo r n c it iz e n s t o t h e e f fe c t s o f
tries, to call for restrictions on migration.
advanced economic integration and migration
policies in other EU countries. In its final and Studies by sector or occupation exist in many
concluding report on the study, the WRR rec- countries but only some of them relate the develop-
ommends that policy should facilitate immi- ment of labour demand to labour supply. These
grants’ integration, but at the same time studies often give an indication of labour market
oblige immigrants to fully participate socially prospects for an occupation or a specific kind of
and economically in all segments of society. education related to that occupation. More recently,
As to labour migration, the WRR recommends kinds of education with good job prospects have
that the government should consider interna- been related to labour market tightness. The very
tional labour migration as a form of interna- limited number of studies that establish a direct link
tional labour participatio n and simplify between emerging labour shortages and the need
bureaucratic procedures. It explicitly recom- for immigrant labour will be discussed below. 113
© OECD 2003Trends in International Migration
1. Sector-level studies on the need for immigrant more youth in IT-oriented studies, but does
workers not give estimates of the extent to which
these shortages could be covered by the
The overview in Annex II.1 of employer associa- domestic labour force. The report apparently
tion’s reports and surveys among employers sug- believes that this is sufficient to cover labour
gested that employers have a pretty good picture of needs because recruiting professionals from
the recruitment difficulties that they can expect in abroad is not among the policy recommenda-
the near future. Economic studies on labour short- tions to meet the future IT labour-market
ages, on the other hand, indicate that precise quan- challenges (USDOC, 1999a and 2000).
titative estimates may not be that straightforward to
give. • In 1999, the German Ministry for Labour and
Social Affairs gave an estimate of 75 000 unfilled
• The Centre for International Economics (CIE), job vacancies in the ICT sector and a growth
commissioned by Australia’s association infor- potential for 250 000 jobs in the 1999-2008
mation and communication technology (ICT) period. It drew consequences for migration
employers, estimates that Australia will expe- policies from this observation and gave the
rience shortage of ICT-professionals. The short- starting sign for a programme to strengthen the
fall of ICT professionals with a university degree supply of ICT specialists. The best-known ele-
is in this study estimated quite precisely at ment of this programme is the “Green-Card”
27 500 persons for the period 2000-2005, over initiative to ease the entry of ICT-specialists
and above the number of graduates and from outside the European Economic Area
migrants that are already expected to join the (EEA). After 11 months, only 8 000 Greencard
industry during that period. Among the rea- visas had been gr an te d, in st ea d of t h e
sons for the shortage the CIE mentions a too planned 20 000 (BMA, 2001). The main draw-
restrictive immigration policy and emigration back of the initiative is seen to be the tempo-
to the United States (CIE, 2001). rary nature of the visas, which are limited to
five years. However, the initiative did yield
• The Technology Administration of the United some benefits. First and foremost, it intensi-
States Department of Commerce has for a fied the discussion on the topic of immigration
number of years drawn attention to the loom- and has contributed to the coming-into-being
ing shortage of information technology (IT) of the Independent Commission on “Immi-
workers (USDOC, 1997a and 1997b, 1999b). gration” (Werner, 2001). Also, being part of a
The Technology Administration has weighed bigger programme, it proved successful in
the views of employers, who believe that raising training places and educational slots
there is not enough trained labour supply, at schools, universities and within compa-
and employees, who believe that labour sup- nies. It is also worth noting that those com-
ply is sufficient both in quality and in quan- panies who did employ foreign IT workers
tity, but that industry does not do enough to with a Green Card showed some important
mobilise existing labour pools. A recent positive spin-offs of immigration. In particular
report describes trends in sector-level labour (BMA, 2001):
supply and demand and discusses signals of
labour market tightness. It anticipates a need – For each Green Card holder, 2.5 comple-
for more than 2 million new IT workers in mentary posts have been created;
the 2000-2008 period, of whom 15% are to
replace retiring and departing workers and – More than four out of five companies report
85% to support new job creation. Of these having obtained a higher competitive stan-
jobs, more than three fourths require at least dard through employing the Green Card
a bachelor’s degree. The report gives signals holder, and nearly half say they have elimi-
of developing labour market tightness and nated capacity problems;
implies important shortages at higher skill
levels. It discusses developing the latent – Nearly one-fifth of all participating compa-
labour supply of women and older workers, nies have renounced plans to relocate a
114 improving the IT-sector’s image and interesting part of their business abroad.
© OECD 2003Labour Shortages and the Need for Immigrants: A Review of Recent Studies
2. Studies on labour shortages by occupation ers at all levels of education and training. Growth
rates are projected to be faster, on average, for occu-
A great number of studies project employment pations requiring at least an associate degree than
growth either two, five or ten years ahead but give for occupations requiring less training. However, the
no indication of labour shortages. They do not anal- largest volume of job creation will be in occupations
yse the development of labour supply and are not requiring less formal education or training, even
linked to immigration policy questions. In general, though many of these occupations are projected to
occupational demand projections estimate future have below-average growth rates. There will also be
replacement and expansion demand: numerous job openings resulting from the need to
• Labour demand as a result from the need to replace workers who leave the labour force or move
replace workers who leave a specific job, to other occupations (Braddock, 1999; see also
switch occupations or leave the labour market Thomson, 2000). Note that BLS has just started with
altogether is a function of the number of peo- a Job Openings and Labour Turnover Survey (JOLTS) to
ple taking retirement, the average age of measure labour market tightness and labour market
(early) retirement, mortality and invalidity. But (matching) efficiency but has not yet analysed the
temporary separations, for example people issue of labour shortages using JOLTS.
taking sabbatical leave, functional mobility To quantify the need for workers in short supply
and emigration, also play a role. If employ- estimates of both the quantity and quality (i.e., the
ment growth is low or nil, labour demand will skill-level), of labour supply, also need to be made.
mostly consist of replacement demand.12 The supply of labour for a specific occupation con-
• The demand for labour to support business sists of first-time labour market entrants, unem-
expansion is a function of economic develop- ployed people and people who move jobs, and
ments, the phase in the economic cycle, eco- suitable people temporarily outside the labour mar-
nomic policy, technological progress and ket. The number of first-time entrants in any forecast
globalisation, among other factors but these period can be worked out by grossing up the num-
factors need not influence the demand in dif- ber of students leaving education in that period
ferent occupations across the labour market and the number of people who have left education
to the same extent. Economic growth in the just before the forecast period and did not gain
last decade has stimulated labour demand in employment.
some sectors more than in other, i.e., ICT and A complicating factor in making accurate prog-
construction. The changing structure of eco- noses about labour supply is the link between edu-
nomic growth, in favour of the tertiary sector, c a ti o n a n d e m pl o y me n t. It i s i m po s si bl e t o
or the above-average growth of specific occu- determine the supply of labour-market entrants on
pations, add an extra dimension and difficulty an occupation-specific level on the basis of the
to the forecasting of labour demand.13 annual number of graduates. To begin with, it is not
Studies of labour demand by occupation can accurate to determine specific levels of qualifica-
predict where employment growth will be the fast- tions from the various types of schools. Some
est. Annex II.2 gives an overview of projections of schools deliver a standard qualification level but for
occupation-level labour demand for 12 OECD coun- other qualifications, notably those required for ICT-
tries. The United States, for example, annually pub- occupations, there are no fixed standards of refer-
lish the Occupational Outlook Handbook. The Office of ence. Another difficulty in the forecasting of first-
Occupational Statistics and Employment Projec- time entrants by occupation or by type of education
tions, under the Department of Labor, develops is that the influence of labour market developments
information about the labour market for the nation and education policy on young peoples’ decisions
as a whole for 10 years in the future. Total employ- to choose a certain occupation or a certain type of
ment is projected to increase by 20.3 million jobs training is almost impossible to model.
over the 1998-2008 period, rising from 140.5 million The most obvious method to estimate future
to 160.8 million, according to the latest projections labour market tightness for first-time entrants is to
of the Bureau of Labor Statistics – (BLS, 2002). work out the development of demand by occupa-
Although employment growth rates will go down tion and to determine which type of education is
compared with those of the previous 10-year period, most wanted for the various occupations. If the
the economy will continue generating jobs for work- demand of first-time entrants with a certain type of 115
© OECD 2003Trends in International Migration
education is greater than the supply, labour market are expected to have in recruiting workers for that
tightness may result. However, a number of things specific occupation.
can obscure the resulting picture:
Applying a similar methodology, the ROA has
• Employers may react to abundant supply by assessed whether the higher education systems in
adju sting th eir re cruitmen t criteria and the EU Member States will provide sufficient num-
employ people who are overqualified (in a bers of Science and Technology (S&T) graduates to
slack labour market) or underqualified (in a meet the demand for scientists and engineers (RSE’s)
tight labour market) for the job.14 up to 2002 (Marley et al., 2001). The report identifies
• Difficulties exist in translating the expected opportunities for international labour mobility to
employment growth by sector in the develop- solve the labour market bottlenecks. The forecasts
ment of total employment at the occupational use four alternative scenarios and indicate consider-
level and the future employment structure by able shortages of RSEs in one or more fields of
qualification. A key of correspondence has to study in all EU Member States except Belgium,
be used, which inevitably results in some Greece, Spain, Finland and the United Kingdom. At
aggregation and hence, a loss of precision. the same time, for the EU as a whole there is excess
One example of a study of the labour market supply in each of the four fields of study distin-
prospects of graduates is the regular employment guished, illustrating the importance of international
projections carried out by the Research Centre for labour mobility.
Education and the Labour Market (ROA), commis- Comparable institutions can be found, for
sioned by the government of the Netherlands (ROA, example, in Australia, Canada and in the United
2001a). The forecasts apply to a total of 207 occupa- Kingdom. The Australian Department of Employ-
tional groups and 104 different education types ment and Workplace Relations publishes the Job
across the labour market. For these groups, the Outlook Report, a guide to the skills projected to be
numbers of people moving into the labour market in demand in the future and highlights occupations
and separating from the labour market are forecast with good prospects. Job Outlook Online is an on-line
for specific future time periods. The demand for version accessible to all interested job-seekers and
labour consists of the demand to replace departing aspirant immigrants (DEWR, 2002). The Australian
workers and the demand to realise business expan- Job Outlook publishes job prospects with a rating.
sion. Labour demand for business expansion is esti- Nearly 400 occupations are listed with an indication
mated on the basis of short and medium term of the career prospects (“Good”, “Average”, “Limited”)
forecasts of the economy by the Netherlands Bureau they offer to first-time labour market entrants. The
for Economic Policy Analysis (CPB). Labour demand prognoses applied to the 2001-2006 period is very
is further distinguished by occupational group and accessible and includes the National Skill Shortage
by type of education. A systematic comparison with List (see Annex II.2). There are two important differ-
labour demand by type of education results in a ences with the Dutch ROA study. First, the DEWR
summary of the labour market perspectives for new pays considerable attention to regional differences
entrants by type of education, and the future possi- in labour market tightness and draws up State skill
ble recruitment difficulties by type of education. shortage lists. Second, there is an explicit link with
The value of the Indicator Future Labour market perspec- immigration policy through:
tives (ITA) indicates for which types of education a
• The facilitating of immigration for people with
future mismatch between supply and demand can
skills that feature on the National Shortage
be expected. Reminiscent of the Australian and
List (more on this in the last section).
Canadian practice describing labour market pros-
pects, the indicator takes the values “Mediocre”, • The Skills Matching Database. Introduced in
“Good”, “Very Good” or “Fair”. Government migration- 1997, the database contains information
policy discussion documents (Tweede Kamer der about skilled workers who have lodged migra-
Staaten Generaal, 2001a and 2001b) and the main tion applications outside Australia in the
advisory bodies to the government (WRR 2001a, Skilled-Independent and Skill-Matching visa
2001b and 2001d), (SER, 2001) frequently quote classes. On the one hand, it is designed to
these results. This is in recognition of the fact that the assist State and Territory governments and
mirror image of good labour market perspectives for employers to attract skilled migrants to areas
116 a certain occupation is the difficulties that employers of Australia who wish to increase their skilled
© OECD 2003You can also read
