The Retirement Prospects of Immigrants: Getting Worse?

Presentation to PMC WinnipegNode Meeting September 29, 2009

Derek Hum

Wayne Simpson

Canada’s Demographic Setting

Aging Population Structure Low Fertility Rates Slow Labour Force Growth

Canada relies on Immigrants

Population Growth Labour Force Growth Productivity

Public Pensions in Canada

Pay As You Go (Entitled/Workers ratio)

Success will depend upon:– Productivity– Labour Force Growth

Plus– Immigration Levels– Immigrant Economic Performance

Pressures on Public Pensions

Increase in ratio of Entitled/Workers

Maintain Benefit Levels Maintain Stable Premiums Permit Flexible Retirement Options Permit Flexible Retirement Date

CPP Benefits related to:

Individuals’ Work History Life time Earnings

Program applies to all Workers ---Native Born and Immigrants

Economic Integration of Immigrants

Do immigrant wages and earnings converge to their Canadian native born counterparts, quickly and completely?

If so, retirement prospects of immigrant are similar to those of other Canadians?

Two Aspects of Convergence

Entry Effect (Typically Negative)

Assimilation Effect (Historically Positive)

Negative entry effect

Immigrants suffer initial disadvantage upon entry to Canada (language, adjustment difficulties, lack of knowledge about labour markets, human capital and social network deficiencies

Immigrants often assumed to possess superior motivation and industry

Assimilation effect (Convergence)

over time, difficulties are overcome, and with superiour motivation and effort, typically “catch up” to native born. (convergence)

In the past, immigrants typically became more “successful” than native born vis a vis earnings

Immigrant Integration (Past)

100

110

120

130

140

150

160

Earn

ing

s

20 30 40 50 60 70 Age(Yrs Since Migration + 20)

Figure 1: Immigrant Integration

native born

immigrant

Immigrant Integration (Present)

100

110

120

130

140

150

160

Earn

ing

s

20 30 40 50 60 70 Age(Yrs Since Migration + 20)

Figure 1: Immigrant Integration

native born

immigrant

Recent Experience of Immigrants

Immigrant Earnings have not converged as quickly

Some estimates indicate growing negative entry effect

Some estimates suggest immigrants will never catch up in their lifetime

Summarizing: Past and Present

Past cohorts experienced small negative entry effects, fast assimilation and rapid convergence (even overtaking native born economic performance)

Present cohorts face: large (and growing) negative entry effects, slow and slowing assimilation, and

Many immigrants will not experience convergence over their working lifetime.

Why Integration has slowed

1/3 decline attributed to source region Declining returns to foreign experience/

schooling for immigrants from “non-traditional” source countries more important

Unemployment effect, especially for immigrants

Convergence and Retirement

Question: Will immigrants achieve earning patterns similar to comparable native born workers?

If so, retirement prospects of immigrants will be similar to native born Canadians.

If not, immigrants will experience lower pensions, higher poverty, lower well being.

Immigrant Earnings Profiles

Typically measured using Census PUMFs, 1981-2001

Cross-sectional PUMFs “blended” to produce “quasi-panel” estimates of immigrant and native born earnings over time

Correct for “across cohort” bias in cross-sectional comparisons to obtain (1) “Entry effect/gap” and (2) “Within cohort” estimates of 5-year growth rates in earnings

Quasi-Panel Estimates of Immigrant Earnings Profiles (Entry and Growth)

Ear

nin

gs

20 30 40 50 60 70

Age

NB

C2

C1

Quasi-Panel Estimates of Immigrant Earnings Profiles (Entry and Growth)

Correct for “cohort bias” arising from “declining cohort quality” of immigrants over time (rising entry effects)

Choose an appropriate native born comparison/control group that “looks like” the foreign born group– 20% Random sample of nb? (Baker&Benjamin,

1994; Grant, 1999; Frenette&Morissette, 2003)

Comparison/control groupof native born:

20% Random sample of nb? Match recent immigrants to nb labour market

entrants (young adults) (Green&Worswick, 2003; Frenette&Morissette, 2003)

Statistical match of the fb sample with the nb sample (nearest neighbour or kernel matching): matching estimator is mean earnings difference between matched samples

Quasi-Panel Estimates of FB Earnings Profiles with Random and Matched NB Samples (B&B Specification)

Immigrant Integration Profiles

-0.5

-0.4

-0.3

-0.2

-0.1

0

0.1

0-5 5-10 10-15 15-20 20-25

Years Since Migration

Gap

with

NB

IM7680

IM8185

IM8690

IM9195

IM7680M

IM8185M

IM8690M

IM9195M

Summary of Results:

Similar results for B&B and F&M specifications

Entry effects are larger with matched NB sample but pattern otherwise similar to random NB sample

Within-cohort growth of earnings substantial and may imply eventual parity/integration

Projections of the immigrant integration profile are unreliable

Implications for Pensionable Earnings or the “Pension Gap”

What are the implications of the gap between fb and nb earnings for private pensions, CPP/QPP?

What “pot of money” at entry would represent the gap faced by an immigrant cohort during a work career of 25 years?

Can calculate for earlier cohorts, e.g. 76-80, and estimate for later cohorts based on “eyeball” estimates of integration profiles

Table 1. Estimated Pension Gapsas Percentage of Native Born

Discount rate IM7680 IM8185 IM8690 IM9195 5.0% 11.4% 17.5% 21.0% 21.7% OLS estimates

(random fb sample) 10.0% 13.1% 20.4% 23.5% 26.3% 5.0% 16.7% 26.4% 26.7% 27.9% Matched estimates

(matched fb sample) 10.0% 17.4% 29.2% 29.3% 32.6%

The Pension Gap in the Survey of Labour and Income Dynamics(work in progress)

SLID 2002 Public File provides tax record information for RPP contributions and private pension income by immigration status, age, and sex

Is there a difference in RPP contributions for fb and nb men who are not retired?

Is there a difference in private pension income for fb and nb men who have retired (older immigrant cohorts)?

Conclusions

Immigrants to Canada continue to do worse in terms of entry earnings relative to nb

Gap in pension incomes between fb and nb men will likely rise

SLID analysis of pension contributions and incomes (to come)